Mechanical and Tribological Properties of Al 2 O 3-TiC Composite Fabricated by Spark Plasma Sintering Process with Metallic ( Ni , Nb ) Binders

Al2O3-10TiC composites were fabricated through the powder metallurgical process (mechanical milling combined with spark plasma sintering) with the addition of Ni/Nb as metallic binders. The effect of binder addition (Ni/Nb) on the processing, microstructure, and mechanical and tribological properties of the bulk-sintered composite samples was investigated. The microstructure of the composite reveals a homogeneous distribution of the TiC particles in the Al2O3 matrix. However, the presence of Ni/Nb was not traceable, owing to the small amounts of Ni/Nb addition. Hardness and density of the composite samples increase with the increasing addition of Nb (up to 2 wt. % Nb). Any further increase in the Nb content (3 wt. %) decreases both the hardness and the wear resistance. However, in case of Ni as binder, both the hardness and wear resistance increases with the increase in the Ni content from 1 wt. % to 3 wt. %. However, the composite samples with Nb as binder show improved hardness and wear resistance compared to the composites with Ni as binder.


Introduction
Ceramic materials are the preferred choice of materials for high temperature, wear resistance and high speed cutting applications because of their high melting points and remarkable hardness [1][2][3].Al 2 O 3 -based ceramic matrix composites are widely used in wear, corrosive, and high temperature environments due to their excellent properties such as high hardness, chemical stability, excellent wear resistance, and high temperature oxidation resistance [4][5][6].However, the wide applications of alumina are limited by their intrinsic brittleness.Hence, enough research has been devoted on ceramic materials to improve their flexural strength and fracture toughness in order to maximize their industrial applications [7][8][9][10].Studies have indicated that the addition of 10 wt.% TiC as a reinforcement to the Al 2 O 3 matrix can improve the hardness, fracture toughness, and thermal shock resistance of the composites [11,12].In addition, introducing ductile reinforcement into the ceramic matrix may effectively improve both the ductility and toughness of the ceramic matrix composites.

Materials and Methods
Al 2 O 3 -10TiC composites were fabricated by SPS process using commercially available α-Al 2 O 3 as matrix (from Alfa Aesar with 99.99% purity), with an average particle size of ~24 µm.TiC particles (from Alfa Aesar with 99.97% purity) with an average particle size of ~7 µm were used as the reinforcement (10 vol.%).In order to improve the bonding between the matrix and reinforcement, Ni & Nb additives (from Himedia, Mumbai, India, 99.99% purity), both with a particle size of ~3 µm, were added with different weight percentages (from 1 to 3 wt.%).The composite powder mixture was prepared by mechanical milling (MM) in a PM-400 planetary ball mill for 30 h in dry condition with a ball to powder ratio of 10:1 and at a set speed of 150 rpm.Phase analysis of the milled powders was carried out using powder diffraction technique with the help of X'Pert PRO PANalytical Diffractometer (from PANalytical, Kassel, Germany) with Cu-kα radiation (λ = 1.54184Å), and the particle size was analyzed using the particle size analyzer (Malvern 2000, New Delhi, India), which is a laser based diffraction technique with water as dispersant.The particle morphologies were observed using field emission scanning electron microscope (FESEM) from JEOL (Freising, Germany) fitted with energy dispersive X-ray spectroscopy (EDX from Oxford Instruments plc, Oxfordshire, UK).The milled powders were loaded inside the SPS system (SPS Syntex, 725, New Delhi, India) with 15 mm diameter top and bottom punches.The sintering was carried out under vacuum with pressure less than 10 −4 MPa.The composite powder was heated to 1673 K at a constant heating rate of 50 K/min and was held for 3 min at a pressure of 60 MPa.
After sintering, the samples were removed from the die to evaluate both the both physical and mechanical properties.The sintered samples were subjected to X-ray diffraction (XRD) after both the surfaces were ground to remove any residual graphite layer present in the sample.The density of the sintered samples was measured using Archimedes principle.Microstructural observation was conducted using the optical microscope (OM from HITACHI, Feldkirchen, Germany) and FESEM.Hardness measurements were carried out using micro-indentation technique with the help of a computer-controlled LECO (AMH43 from LECO, Mönchengladbach, Germany) Vicker's microhardness tester.The micro-hardness device was equipped with a typical diamond indenter in the form of a pyramid with square base and an angle of 136 • between the opposite faces.Indentations were carried out with an applied load of 1 g and a dwell time of 13 s.The abrasion test was performed using a high temperature abrasion tester-CM-9101 (according to ASTM G65 standards) at 873 K for 5 min cycling time under the applied load of 5 kg.These tests were carried out to analyze the weight loss and abrasion rate at given temperature and cycling time.
The CM-9101 abrasion tester set-up has a rotating metallic (Ni-based Inconel alloy) wheel of 100 mm diameter, which acts as a counter face against abrasive sample surface.The specimens were weighed initially on an electronic weighing machine, with an accuracy of 0.1 mg.They were then fixed in the specimen holder, which is located at one end of the lever, and the door was closed.The samples were heated to 873 K inside the chamber as per the working requirement of the cutting tool insert.5 kg load was applied on the other end of the hydraulic arm with a rotation speed of 200 rpm, which is maintained by adjusting the knob.White alumina sand AFS 50/70 (with a hardness of 2000 Hv) as abrasive media was loaded in the hopper of the machine.The abrasive medium falls at a rate of 350 g/min between the specimen and the counter face during the entire run time of the experiment.The applied load presses the specimen against the counter face, while there is a continuous flow of abrasive medium between the specimen and the counter face.All the experiments were conducted for a constant time of 5 min.After completion of the experiment, the samples were again weighed and weight loss was recorded.The abrasion wear rate was calculated using the formula [38], where W a is the abrasion wear rate (mm 3 /Nm), ∆G is mass loss (g), d is density (g/mm 3 ), M is the applied load (N), and S is the sliding distance (m) [39,40].

Power Characterization
XRD patterns of the milled Al 2 O 3 -10TiC powder samples with and without the addition of binder are shown in Figure 1.The XRD patterns show the presence of only two phases with no new phases formed, indicating that no reaction has taken place between the matrix, reinforcement, and the binder during MM.In addition, the patterns do not show the presence of Ni/Nb peaks, which may be due to the low contents of Ni/Nb (<5 wt.%), which are below the detection limit of the XRD device.Hence, all the identified peaks correspond to the two phases Al 2 O 3 and TiC. Figure 2a,c shows the FESEM micrograph of the Al 2 O 3 -10TiC milled powder (with 3 wt.% nickel) and their size distribution.It show two types of particles, big polygon-shaped Al 2 O 3 particles (8-10 µm) and small equiaxed TiC particles (1-2 µm), distributed uniformly throughout the matrix.The presence of Ni/Nb was hard to identify due to the low amount of binder addition in the composite.Figure 2b,d reveals the Gaussian distribution function of the powder particles before and after MM.It shows that the particle size is reduced after MM and most of the particles fall between the range 1-10 µm after MM, as compared to 1-20 µm in the un-milled condition.The results are representative, since the size of the particles remain similar irrespective of the amount and type (Ni/Nb) of binder added and hence not shown here.This may be that Ni/Nb does not react with the matrix and/or the reinforcement.In addition, both these elements (Ni/Nb) have similar hardness levels (between 4.0 and 6.0) on Mohs scale.

Microstructure and Mechanical Property of Bulk Sample
The FESEM micrographs of the bulk composite samples with and without the addition of different weight percentage (1-3 wt.%) binder (Ni/Nb) are shown in Figure 3a-g.The micrograph revels that the composite samples consist of two phases (bright and dark) distributed homogeneously in all the measured samples.It has been observed from the EDX analysis that the bright phase corresponds to TiC particles and the dark phases represents Al2O3 particles (see Figure 4).Traces of Ni/Nb are hard to be observed because of the low contents and the size of the particles.The micrographs does not show the presence of defects like porosity in the sample, suggesting these samples have a very high density.In addition, the microstructures also reveal the presence of an excellent interfacial bonding between the matrix (Al2O3) and the reinforcement (TiC) without the presence of any interfacial defects between the matrix and the reinforcement.

Microstructure and Mechanical Property of Bulk Sample
The FESEM micrographs of the bulk composite samples with and without the addition of different weight percentage (1-3 wt.%) binder (Ni/Nb) are shown in Figure 3a-g.The micrograph revels that the composite samples consist of two phases (bright and dark) distributed homogeneously in all the measured samples.It has been observed from the EDX analysis that the bright phase corresponds to TiC particles and the dark phases represents Al2O3 particles (see Figure 4).Traces of Ni/Nb are hard to be observed because of the low contents and the size of the particles.The micrographs does not show the presence of defects like porosity in the sample, suggesting these samples have a very high density.In addition, the microstructures also reveal the presence of an excellent interfacial bonding between the matrix (Al2O3) and the reinforcement (TiC) without the presence of any interfacial defects between the matrix and the reinforcement.

Microstructure and Mechanical Property of Bulk Sample
The FESEM micrographs of the bulk composite samples with and without the addition of different weight percentage (1-3 wt.%) binder (Ni/Nb) are shown in Figure 3a-g.The micrograph revels that the composite samples consist of two phases (bright and dark) distributed homogeneously in all the measured samples.It has been observed from the EDX analysis that the bright phase corresponds to TiC particles and the dark phases represents Al 2 O 3 particles (see Figure 4).Traces of Ni/Nb are hard to be observed because of the low contents and the size of the particles.The micrographs does not show the presence of defects like porosity in the sample, suggesting these samples have a very high density.In addition, the microstructures also reveal the presence of an excellent interfacial bonding between the matrix (Al 2 O 3 ) and the reinforcement (TiC) without the presence of any interfacial defects between the matrix and the reinforcement.The densification and hardness of the composites with the binder addition (Ni/Nb) are shown in Figure 5.The density of the composites increase from 3.9820 g/cc to 4.0713 g/cc, when the amount of Nb addition is increased from 1 wt.% to 2 wt.%, which amount to an increase in the relative density from ~96.5% to ~98%.The relative density shows a marginal decline (decreases to ~97.5%), when the Nb addition is increased to 3 wt.%.This marginal decrease in the relative density of the 3 wt.% Nb added composite may be attributed to the difference in wettability levels.Nb does not have a good wettability with the Al2O3 matrix as compared to Ni, and any increase in the Nb content beyond 2 wt.% does not help to improve the density of the Al2O3-10TiC composites.Similar trend is observed with the hardness data as well.The hardness levels of the composites increase with increasing addition of Nb from ~1410 MPa (without Nb addition) to ~1680 MPa for 1 wt.% addition of Nb.With further addition of Nb to 2 wt.%, the hardness on the composite increases to ~1870 MPa and then marginally decreases to 1770 MPa with 3 wt.% Nb addition.This decline in the hardness is attributed to the decrease in the relative density of the composite, which again is related to the wettability of Nb with Al2O3.
On the other hand, with the addition of Ni, both the density and the relative densities increase linearly.For instance, the relative density of the composites increases from ~96% (1 wt.% Ni addition) The densification and hardness of the composites with the binder addition (Ni/Nb) are shown in Figure 5.The density of the composites increase from 3.9820 g/cc to 4.0713 g/cc, when the amount of Nb addition is increased from 1 wt.% to 2 wt.%, which amount to an increase in the relative density from ~96.5% to ~98%.The relative density shows a marginal decline (decreases to ~97.5%), when the Nb addition is increased to 3 wt.%.This marginal decrease in the relative density of the 3 wt.% Nb added composite may be attributed to the difference in wettability levels.Nb does not have a good wettability with the Al 2 O 3 matrix as compared to Ni, and any increase in the Nb content beyond 2 wt.% does not help to improve the density of the Al 2 O 3 -10TiC composites.Similar trend is observed with the hardness data as well.The hardness levels of the composites increase with increasing addition of Nb from ~1410 MPa (without Nb addition) to ~1680 MPa for 1 wt.% addition of Nb.With further addition of Nb to 2 wt.%, the hardness on the composite increases to ~1870 MPa and then marginally decreases to 1770 MPa with 3 wt.% Nb addition.This decline in the hardness is attributed to the decrease in the relative density of the composite, which again is related to the wettability of Nb with Al 2 O 3 .On the other hand, with the addition of Ni, both the density and the relative densities increase linearly.For instance, the relative density of the composites increases from ~96% (1 wt.% Ni addition) to ~96.5 wt.% (2 wt.% Ni) and finally to ~97% for 3 wt.% Ni addition.Similar trend has been observed also for the hardness values, where the hardness increases from ~1410 MPa without Ni addition to ~1625 MPa, ~1780 MPa, and ~1800 MPa for the addition of 1 wt.%, 2 wt.%, and 3 wt.% Ni, respectively.Such consistent increase in both relative densities and hardness of the composites with the addition of Ni is attributed to the complete wettability of Ni with the matrix Al 2 O 3 .It has been observed that the complete wettability provides better bonding strength with matrix and helps to reduce the amount of porosity, which results in the enhanced densities, flexural strength, and fracture toughness [41][42][43].A highest hardness of ~1870 MPa is observed for the composites with 2 wt.% Nb addition and a least hardness of ~1625 MPa is observed for the composite with 1 wt.% Ni addition.In general, the samples with the addition of Nb shows better strength levels (hardness) than the composites with Ni addition.This may be attributed to the strength of Nb, which is slightly higher (Mohs scale 6.0) compared to Ni (Mohs scale 4.0).
Metals 2018, 8, 50 6 of 11 to ~96.5 wt.% (2 wt.% Ni) and finally to ~97% for 3 wt.% Ni addition.Similar trend has been observed also for the hardness values, where the hardness increases from ~1410 MPa without Ni addition to ~1625 MPa, ~1780 MPa, and ~1800 MPa for the addition of 1 wt.%, 2 wt.%, and 3 wt.% Ni, respectively.Such consistent increase in both relative densities and hardness of the composites with the addition of Ni is attributed to the complete wettability of Ni with the matrix Al2O3.It has been observed that the complete wettability provides better bonding strength with matrix and helps to reduce the amount of porosity, which results in the enhanced densities, flexural strength, and fracture toughness [41][42][43].A highest hardness of ~1870 MPa is observed for the composites with 2 wt.% Nb addition and a least hardness of ~1625 MPa is observed for the composite with 1 wt.% Ni addition.
In general, the samples with the addition of Nb shows better strength levels (hardness) than the composites with Ni addition.This may be attributed to the strength of Nb, which is slightly higher (Mohs scale 6.0) compared to Ni (Mohs scale 4.0).

Abrasive Wear Study
Ceramic materials possess high hardness and relative low fracture toughness [44,45], and hence brittle micro-cutting wear mechanisms will predominate.However, sometimes plastic deformation may also be observed during an abrasion wear process [46].Figures 6 and 7 show the microstructure of the surfaces before and after abrasion test with the addition of Nb and Ni binders, respectively.Generally, from these images, it can be observed that the abrasive surfaces do not show any deep cutting mark, which is due to the presence of hard particles on the moving path of abradants.These hard particles are capable of resisting the abradant, resulting either in the rolling of abradants or the to ~96.5 wt.% (2 wt.% Ni) and finally to ~97% for 3 wt.% Ni addition.Similar trend has been observed also for the hardness values, where the hardness increases from ~1410 MPa without Ni addition to ~1625 MPa, ~1780 MPa, and ~1800 MPa for the addition of 1 wt.%, 2 wt.%, and 3 wt.% Ni, respectively.Such consistent increase in both relative densities and hardness of the composites with the addition of Ni is attributed to the complete wettability of Ni with the matrix Al2O3.It has been observed that the complete wettability provides better bonding strength with matrix and helps to reduce the amount of porosity, which results in the enhanced densities, flexural strength, and fracture toughness [41][42][43].A highest hardness of ~1870 MPa is observed for the composites with 2 wt.% Nb addition and a least hardness of ~1625 MPa is observed for the composite with 1 wt.% Ni addition.
In general, the samples with the addition of Nb shows better strength levels (hardness) than the composites with Ni addition.This may be attributed to the strength of Nb, which is slightly higher (Mohs scale 6.0) compared to Ni (Mohs scale 4.0).

Abrasive Wear Study
Ceramic materials possess high hardness and relative low fracture toughness [44,45], and hence brittle micro-cutting wear mechanisms will predominate.However, sometimes plastic deformation may also be observed during an abrasion wear process [46].Figures 6 and 7 show the microstructure of the surfaces before and after abrasion test with the addition of Nb and Ni binders, respectively.Generally, from these images, it can be observed that the abrasive surfaces do not show any deep cutting mark, which is due to the presence of hard particles on the moving path of abradants.These

Abrasive Wear Study
Ceramic materials possess high hardness and relative low fracture toughness [44,45], and hence brittle micro-cutting wear mechanisms will predominate.However, sometimes plastic deformation Metals 2018, 8, 50 7 of 12 may also be observed during an abrasion wear process [46].Figures 6 and 7 show the microstructure of the surfaces before and after abrasion test with the addition of Nb and Ni binders, respectively.Generally, from these images, it can be observed that the abrasive surfaces do not show any deep cutting mark, which is due to the presence of hard particles on the moving path of abradants.These hard particles are capable of resisting the abradant, resulting either in the rolling of abradants or the pulling out of particles themselves.Locally, some surfaces become uneven and even show the erosion of some fine particles.However, a uniform wear throughout the sample surface is observed in most of the samples.In the Nb added composites, some micro-cracks and detachment of the particles are observed in samples especially at higher amounts of Nb (3 wt.% Nb) addition, as marked by arrows in the Figure 6h.However, the presence of micro-cracks and detachment of particles is comparatively less in the samples with 1 wt.% and 2 wt.% addition of Nb.This effect is attributed to the decrease in the relative density and, in turn, hardness in the composites with 3 wt.% Nb addition, compared to the other two samples with 1 wt.% and 2 wt.% Nb addition, respectively.In case of the composites with Ni as binder, samples surface changes show a uniform trend, which is in accordance with the increase in the addition of Ni content.
pulling out of particles themselves.Locally, some surfaces become uneven and even show the erosion of some fine particles.However, a uniform wear throughout the sample surface is observed in most of the samples.In the Nb added composites, some micro-cracks and detachment of the particles are observed in samples especially at higher amounts of Nb (3 wt.% Nb) addition, as marked by arrows in the Figure 6h.However, the presence of micro-cracks and detachment of particles is comparatively less in the samples with 1 wt.% and 2 wt.% addition of Nb.This effect is attributed to the decrease in the relative density and, in turn, hardness in the composites with 3 wt.% Nb addition, compared to the other two samples with 1 wt.% and 2 wt.% Nb addition, respectively.In case of the composites with Ni as binder, samples surface changes show a uniform trend, which is in accordance with the increase in the addition of Ni content.addition until 2 wt.%.Addition of Nb content beyond 2 wt.% decreases both the hardness and wear resistance, due to issues with the wettability, and hence there is a decrease in the relative density of the composites.However, with the addition of Ni, both the hardness and wear resistance increases with increase in the Ni content.In general, the hardness and wear resistance of the composites are superior when Nb is added as a binder rather than Ni.This may be attributed to the marginal increase in hardness of Nb compared to Ni.The weight loss and the wear rate data of the composites with the addition of Nb and Ni are shown in Figure 8.Both the weight loss and wear rate data show that the addition of both Nb and Ni in general improves the wear resistance of the material, due to improved relative densities and, in turn, the hardness of the composites.However, the lone exception is the composite sample with 3 wt.% Nb addition, where weight loss increases compared to the sample with 2 wt.% Nb addition, suggestion poor wear resistance.The results are in compliance with the FESEM images and also with the relative density and hardness data.From these experimental data, the following observations can be made: both the hardness (strength) and the wear resistance increases with increase in the Nb addition until 2 wt.%.Addition of Nb content beyond 2 wt.% decreases both the hardness and wear resistance, due to issues with the wettability, and hence there is a decrease in the relative density of the composites.However, with the addition of Ni, both the hardness and wear resistance increases with increase in the Ni content.In general, the hardness and wear resistance of the composites are superior when Nb is added as a binder rather than Ni.This may be attributed to the marginal increase in hardness of Nb compared to Ni.

Summary
Al2O3-10TiC bulk ceramic matrix composites were successfully prepared by powder metallurgical route (mechanical milling combined with spark plasma sintering) with the addition of Ni/Nb as metallic additives.
 XRD data shows the presence of two phases Al2O3 and TiC, confirming no reaction between the matrix, reinforcement, and metallic binders during the mechanical milling and subsequent spark plasma sintering processes.However, no peaks of Ni/Nb are observed owing to the small amounts of Ni/Nb addition, which is below the deductible limits of the XRD setup. The microstructure of the composite reveals homogeneous distribution of the TiC particles within the Al2O3 matrix.The relative density, hardness, and the wear resistance of the composites increase with the addition of Nb until 2 wt.%.Any addition of Nb beyond 2 wt.% (3 wt.% addition of Nb) hampers their properties, due to relatively poor wettability of Nb with the Al2O3 matrix.Nevertheless, the properties of composites with 3 wt.% addition of Nb are better than the samples without the addition of Nb.  On the other hand, with the addition of Ni, the relative density, hardness, and wear resistance increases with increase in the Ni content.In general, the samples with the addition of Nb as binder show more improved properties than the composites with Ni addition, which is attributed to the marginal increase in the hardness of Nb compared to Ni.  Overall results show that the addition of metallic binders improve the relative density, hardness, and wear resistance of the Al2O3-10TiC composite samples.

Summary
Al 2 O 3 -10TiC bulk ceramic matrix composites were successfully prepared by powder metallurgical route (mechanical milling combined with spark plasma sintering) with the addition of Ni/Nb as metallic additives.
XRD data shows the presence of two phases Al 2 O 3 and TiC, confirming no reaction between the matrix, reinforcement, and metallic binders during the mechanical milling and subsequent spark plasma sintering processes.However, no peaks of Ni/Nb are observed owing to the small amounts of Ni/Nb addition, which is below the deductible limits of the XRD setup.The microstructure of the composite reveals homogeneous distribution of the TiC particles within the Al 2 O 3 matrix.The relative density, hardness, and the wear resistance of the composites increase with the addition of Nb until 2 wt.%.Any addition of Nb beyond 2 wt.% (3 wt.% addition of Nb) hampers their properties, due to relatively poor wettability of Nb with the Al 2 O 3 matrix.Nevertheless, the properties of composites with 3 wt.% addition of Nb are better than the samples without the addition of Nb.On the other hand, with the addition of Ni, the relative density, hardness, and wear resistance increases with increase in the Ni content.In general, the samples with the addition of Nb as binder show more improved properties than the composites with Ni addition, which is attributed to the marginal increase in the hardness of Nb compared to Ni. Overall results show that the addition of metallic binders improve the relative density, hardness, and wear resistance of the Al 2 O 3 -10TiC composite samples.

Figure 1 .
Figure 1.XRD (Cu-Kα) patterns of the Al2O3-10TiC-milled powder with the different binder content with (a) Ni and (b) Nb addition.

Figure 5 .
Figure 5. Densification and hardness plot of the Al2O3-TiC composite samples with different wt.% of (a) Nb and (b) Ni binder addition.

Figure 5 .
Figure 5. Densification and hardness plot of the Al2O3-TiC composite samples with different wt.% of (a) Nb and (b) Ni binder addition.

Figure 5 .
Figure 5. Densification and hardness plot of the Al 2 O 3 -TiC composite samples with different wt.% of (a) Nb and (b) Ni binder addition.

Figure 6 .
Figure 6.FESEM micrographs of the Al2O3-10TiC composite samples without Nb and with different contents of Nb (a-d) before and (e-h) after abrasion test.The weight loss and the wear rate data of the composites with the addition of Nb and Ni are shown in Figure8.Both the weight loss and wear rate data show that the addition of both Nb and Ni in general improves the wear resistance of the material, due to improved relative densities and, in turn, the hardness of the composites.However, the lone exception is the composite sample with 3 wt.

Figure 6 .
Figure 6.FESEM micrographs of the Al 2 O 3 -10TiC composite samples without Nb and with different contents of Nb (a-d) before and (e-h) after abrasion test.

Figure 7 .
Figure 7. FESEM micrographs of the Al2O3-10TiC composite samples without Ni and with different contents of Ni (a-d) before and (e-h) after abrasion test.

Figure 7 .
Figure 7. FESEM micrographs of the Al 2 O 3 -10TiC composite samples without Ni and with different contents of Ni (a-d) before and (e-h) after abrasion test.

Figure 8 .
Figure 8. Plots showing the effect of Nb and Ni binder on the (a) weight loss and (b) abrasion rate on Al2O3-10TiC composite samples.

Figure 8 .
Figure 8. Plots showing the effect of Nb and Ni binder on the (a) weight loss and (b) abrasion rate on Al 2 O 3 -10TiC composite samples.