MATERIALS TRANSACTIONS
Online ISSN : 1347-5320
Print ISSN : 1345-9678
ISSN-L : 1345-9678
Effects of the Sintering Conditions on the Mechanical Properties of Titanium-Carbide-Particle-Reinforced Magnesium Nanocomposites Fabricated by Mechanical Alloying/Mechanical Milling/Spark Plasma Sintering
Shigehiro KawamoriYoshihumi KawashimaHiroshi FujiwaraKiyoshi KurodaYukio Kasuga
Author information
JOURNALS FREE ACCESS FULL-TEXT HTML

2018 Volume 59 Issue 1 Pages 82-87

Details
Abstract

To enhance the mechanical properties of Mg alloys, we have fabricated Mg/TiC composites by reinforcing the Mg matrix composed of nanosize crystal grains with 20 vol% TiC nanoparticles. The Mg/TiC nanocomposites were fabricated by mechanical milling (MM) and spark plasma sintering (SPS). The TiC nanoparticles were produced by mechanical alloying (MA). The effects of the applied pressure and holding time during SPS on the mechanical properties of this nanocomposite were investigated. Microstructure observations and elemental analysis show that the TiC particles (TiCp) in the nanocomposites have an ultrafine microstructure with an average particle size of approximately 9 nm and they aggregate within the Mg matrix. The Vickers hardness of the nanocomposites increases to 150 HV when the SPS applied pressure and holding time are increased. However, the increase in the hardness is accompanied by a decrease in the bending strength. The main factors for the improvement of the mechanical properties of the 20 vol% TiCp/Mg nanocomposite are considered to be the density and compressive residual stress.

1. Introduction

In recent years, advances in automotive weight reduction have led to increased demand for Mg alloys as lightweight substitutes for Al alloys. However, Mg alloys generally have inferior mechanical properties to Al alloys, including lower hardness, 0.2% yield stress, tensile strength, and bending strength. To improve these mechanical properties, composites of pure Mg reinforced by ceramic particles have been fabricated using powder metallurgical processes, and their mechanical properties have been investigated.17)

Titanium carbide (TiC) has desirable properties as a reinforcing material in composites. TiC has higher hardness (3,200 HV),8) elastic modulus (451 GPa),8) wear resistance, and thermal stability than other popular reinforcing materials, such as Al2O3 and SiC. For ceramic-particle-reinforced metal matrix composites, the most important factor is the bonding strength of the interface between the metal and the ceramic particles, which is generally evaluated by the wettability between the ceramic and the metal. Al2O3 and SiC have relatively good wettability with magnesium.9,10). However, the wettability between TiC and Mg is considered to be poorer than that between SiC and Mg because TiC has higher chemical stability than SiC.11,12) Because the surface of TiC becomes active and its specific surface area becomes large when in the form of nanoparticles, improvement of bondability between TiC and Mg is expected for TiC nanoparticles.

In this study, to achieve good bondability between Mg and TiC particles (TiCp) and improve the mechanical properties of TiCp/Mg composites, a Mg matrix composite with nanosize crystal grains reinforced by TiC nanoparticles was fabricated by long duration milling treatment of pure Mg powder and TiC nanoparticles by mechanical milling (MM) and spark plasma sintering (SPS). The TiC nanoparticles were produced by long duration milling treatment of Ti and C powders by the mechanical alloying (MA). Finally, the effects of the applied pressure and holding time during SPS on the mechanical properties of TiCp/Mg nanocomposites were investigated.

2. Experimental Procedure

2.1 Fabrication of TiCp/Mg nanocomposites

Figure 1 shows the MA process for producing TiC nanoparticles from Ti and C powders. The starting materials were pure Ti powder (purity 99.9%, 45 μm particle size, Kojundo Chemical Laboratory Co., Ltd.) and C powder (purity 99.7%, 5 μm particle size, Kojundo Chemical Laboratory Co., Ltd.). Mixtures of the Ti powder and 50 mol% C powder (20 cm3) were fed into a ZrO2 container with a capacity of 250 cm3 together with 5-mm-diameter ZrO2 balls (80 cm3) in an Ar atmosphere. Stearic acid (8 mass%) was added as a lubricant. This concentration of stearic acid was used because it is the lower limit to prevent the MA and MM powders from adhering to the ball mill container. To mill the mixed powders, the ZrO2 container was rotated at 400 rpm for 518.4 ks in an Ar atmosphere by a planetary ball mill (P-5, Fritsch GmbH).

Fig. 1

MA process for production of TiCp from Ti and C powders.

Figure 2 shows the MM/SPS process for fabricating the 20 vol% TiCp/Mg nanocomposites. Pure Mg powder with a particle size of 180 μm (purity 99.5%, Kojundo Chemical Laboratory Co., Ltd.) and TiCp obtained by the above MA process were mixed at a volume ratio of 8:2. The mixture of the Mg and TiCp powders (40 cm3) was fed into a ZrO2 container with a capacity of 250 cm3 together with 5-mm-diameter ZrO2 balls (80 cm3) in an Ar atmosphere. Stearic acid (8 mass%) was added as a lubricant. To mill the mixed powders, the ZrO2 container was rotated at 400 rpm for 259.2 ks in an Ar atmosphere by the planetary ball mill.

Fig. 2

MM/SPS process to fabricate the 20 vol% TiCp/Mg nanocomposites.

TiCp/Mg composites were obtained with many different combinations of the MA and MM conditions (ball mill rotation number and treatment time) under the same SPS conditions. The MA/MM conditions that gave the TiCp/Mg composite with the highest density and Vickers hardness were used as the ball mill rotation number and treatment time.

The 20 vol% TiCp/Mg powders were densified in a graphite die (20 mm inner diameter) using a SPS machine (Dr. Sinter Lab SPS-515S, Fuji Electronic Industrial Co., Ltd.) at 40–80 MPa and 848 K for 0.6 or 3.6 ks in an Ar atmosphere. The graphite die was then cooled to below 323 K in the SPS machine to obtain the 20 vol% TiCp/Mg nanocomposite.

2.2 Characterization

X-ray diffraction (XRD) (RAD-IIIB, Rigaku Co., Ltd.) was performed to identify the constituent phases in the TiC nanoparticles of the 20 vol% TiCp/Mg powders and 20 vol% TiCp/Mg nanocomposites. Residual stress evaluation of the TiCp/Mg nanocomposites was performed using a micro-area X-ray residual stress measurement system (AutoMATE II; Rigaku Co., Ltd.). Microstructural observation and elemental analysis of the TiCp/Mg nanocomposites were performed by transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDS) (JEM-2100F and JED-2300T, JEOL Co., Ltd.), respectively.

In the elemental analysis of the TiCp/Mg nanocomposites, mapping and point analysis were performed by scanning transmission electron microscopy (STEM)–EDS. The point analysis results are “semiquantitative analysis” values obtained from the correction value of the characteristic X-ray intensities of each element without using the standard material to obtain the approximate element composition of the components in the nanocomposites. Identification of the components was performed by combining the results with the XRD results.

The densities of the TiCp/Mg nanocomposites were measured by a densimeter (DME-220, Shinko Denshi Co., Ltd.). The surface hardness values of the TiCp/Mg nanocomposites were measured by a Vickers hardness tester (MV-1, Mori Testing Machine Co., Ltd.) at 49 N for 10 s. The bending specimens were obtained by machining the disc-shaped TiCp/Mg nanocomposites to a size of 19.6 mm (length) × 5.0 mm (width) × 1.4 mm (thickness) using an electric discharge machine (A500W, Sodick Co., Ltd.). After buffing both surfaces of the specimens, three-point bending tests were performed at a crosshead speed of 2.0 mm/min using a 20-kN hydraulic universal testing machine (SC-20CJ, Tokyo Koki Testing Machine Co., Ltd.). In the bending tests, the maximum loads of five specimens were measured for each SPS condition. The bending strengths were calculated from the obtained maximum loads using the general equation.

3. Results and Discussion

3.1 Microstructures of the TiCp/Mg nanocomposites

Figure 3(a)–(f) show the XRD patterns of the MA-processed Ti and C powders for MA processing times of 10.8, 21.6, 32.4, 43.2, 86.4, and 518.4 ks, respectively. The results show that TiC begins to be produced by the solid-state reaction of Ti and C after 21.6 ks of processing, and by 86.4 ks only TiC is present. In this study, TiCp used for nanocomposite preparation was produced using the longest MA processing time (518.4 ks).

Fig. 3

XRD patterns of the Ti and C MA powders for MA times of 10.8 (a), 21.6 (b), 32.4 (c), 43.2 (d), 86.4 (e), and 518.4 ks (f).

Figure 4(a)–(c) show the XRD patterns of the 20 vol% TiCp/Mg nanocomposites after SPS processing with SPS applied pressures and holding times of 40 MPa and 0.6 ks, 80 MPa and 0.6 ks, and 80 MPa and 3.6 ks, respectively. The nanocomposites obtained with all of the conditions show the same constituent phases. In addition to Mg and TiC, both MgO and ZrO2 are present in the nanocomposites. It is considered that MgO is present because of oxidation of Mg owing to the temperature increase in the ball mill container and carbon die during MM/SPS processing and ZrO2 is present because of contamination from the ZrO2 balls and container in MA/MM processing.

Fig. 4

XRD patterns of the 20 vol% TiCp/Mg nanocomposites fabricated with SPS applied pressures and holding times of 40 MPa and 0.6 ks (a), 80 MPa and 0.6 ks (b), and 80 MPa and 3.6 ks (c).

Figure 5(a) shows a TEM image of the nanocomposite prepared with an applied pressure of 80 MPa, a sintering temperature of 848 K, and a holding time of 3.6 ks. The Mg, O, Ti, and C elemental analysis results of the nanocomposite are shown in Fig. 5(b)–(d), respectively. The TEM image shows many ultrafine particles with particle sizes of 4–19 nm (average particle size ~9 nm) and the particles are aggregated. From the elemental analysis results, the particles are presumed to be TiC. However, while Ti is clearly present in these particles, it is difficult to determine whether they also contain C.

Fig. 5

TEM image (a) and elemental analysis of Mg (b), O (c), Ti (d), and C (e) for the 20 vol% TiCp/Mg nanocomposite produced with a SPS applied pressure of 80 MPa and a holding time of 3.6 ks.

Semiquantitative analysis of several particles was performed for the TiCp/Mg nanocomposite produced with an applied pressure of 80 MPa, a sintering temperature of 848 K, and a holding time of 3.6 ks. Figure 6(a) and (b) show a TEM image and the semiquantitative analysis results for the location points shown in Fig. 6(a), respectively. It is possible that TiC and C coexist in TiCp fabricated with a stoichiometric ratio of Ti to C because the composition of Ti in the TiC phase is 50–68 mol% from the Ti–C equilibrium diagram.13) Therefore, the particle at point 2 is TiC and the particles at points 1 and 3 are both presumed to be TiC and C. The presence of Mg and O in the semiquantitative analysis results is because of incorporation of Mg and O located close to the particles.

Fig. 6

TEM image of the 20 vol% TiCp/Mg nanocomposite produced with a SPS applied pressure of 80 MPa and holding time of 3.6 ks (a) and quantitative analysis results of the particles shown in (a) (b).

Figure 7(a) and (b) show a TEM image and the semiquantitative analysis results for the location points shown in Fig. 7(a), respectively, for the TiCp/Mg nanocomposite produced with an applied pressure of 80 MPa, a sintering temperature of 848 K, and a holding time of 0.6 ks. The particles at points 4 and 5 have a size of 10 nm and are identified as MgO. ZrO2 identified by XRD could not be confirmed by TEM observation or elemental analysis.

Fig. 7

TEM image of the 20 vol% TiCp/Mg nanocomposite produced with a SPS applied pressure of 80 MPa and a holding time of 0.6 ks (a) and quantitative analysis results of the particles shown in (a) (b).

The TEM observations of the samples fabricated under different SPS conditions show that the grain sizes of the crystal grains identified as Mg solid solutions are 8–11 nm regardless of the SPS conditions. The effects of the SPS conditions on the microstructure of the TiCp/Mg nanocomposite, such as the size, shape, composition, and dispersion state of the TiC particles, MgO particles, and Mg grains, are not clear from the results of XRD, TEM observation, and elemental analysis using STEM–EDS.

3.2 Densities of the TiCp/Mg nanocomposites

Figure 8 shows the effects of the SPS applied pressure and holding time on the density of the TiCp/Mg nanocomposite. As the applied pressure increases, the density only slightly changes at a holding time of 0.6 ks, but it increases at a holding time of 3.6 ks. The reason for the density increase at the longer SPS holding time is thought to be because of further progress of densification rather than microstructural changes, because the different SPS conditions appear to have little effect on the nanocomposite microstructure, as discussed above.

Fig. 8

Effects of the SPS applied pressure and holding time on the density of the 20 vol% TiCp/Mg nanocomposite.

3.3 Mechanical properties of the TiCp/Mg nanocomposites

Figure 9 shows the effects of the SPS applied pressure and holding time on the Vickers hardness of the TiCp/Mg nanocomposite. The Vickers hardness increases as the applied pressure increases, especially for the longer holding time (3.6 ks). In addition, the variation of the hardness for a holding time of 0.6 ks is larger than that for a holding time of 3.6 ks. The sample processed at an applied pressure of 80 MPa and a holding time of 3.6 ks has the highest hardness value of approximately 150 HV, which is much higher than that of the high-strength AZ91 Mg alloy.14) From Fig. 8, as the applied pressure increases, the density hardly changes for the short holding time, while it increases for the long holding time. For a long holding time, we believe that the hardness increases with increasing applied pressure because of densification of the nanocomposite.

Fig. 9

Effects of the SPS applied pressure and holding time on the Vickers hardness of the 20 vol% TiCp/Mg nanocomposite. The hardness range of the AZ91 Mg alloy is shown for comparison.

Figure 10 shows the effects of the SPS applied pressure and holding time on the compressive residual stress of the TiCp/Mg nanocomposite. Regardless of the holding time, the compressive residual stresses increases with increased applied pressure. We believe that only the residual compressive strain increases with increasing SPS applied pressure because the change in the microstructure of the nanocomposite accompanying the increase in the SPS applied pressure is not large from the results of microstructure observation and analysis. Therefore, we conclude that the compressive residual stress is responsible for the increase in the hardness with increasing applied pressure, regardless of the holding time. Furthermore, we believe that the density shown in Fig. 8 is influenced by the increase in the slope of the hardness curve against the applied pressure as the holding time increases.

Fig. 10

Effects of the SPS applied pressure and holding time on the compressive residual stress of the 20 vol% TiCp/Mg nanocomposite.

Figure 11 shows the effects of the SPS applied pressure and holding time on the bending strength of the TiCp/Mg nanocomposite. Regardless of the holding time, the bending strength tends to decrease with increasing pressure. In addition, for the longer holding time (3.6 ks), the bending strength variation decreases compares the variation for the shorter holding time (0.6 ks).

Fig. 11

Effects of the SPS applied pressure and holding time on the bending strength of the 20 vol% TiCp/Mg nanocomposite.

Figure 12 shows scanning electron microscope (SEM) images of the indentations produced by Vickers hardness testing of TiCp/Mg nanocomposites fabricated with SPS pressures of 40 (Fig. 12(a) and (b)) and 80 MPa (Fig. 12(c) and (d)) for a holding time of 3.6 ks. Cracks formed at the indentation corners in the sample processed at 80 MPa. We believe that because the compressive residual stress in the nanocomposite increases with increasing SPS applied pressure (Fig. 10), the hardness increases and the ductility decreases. When indenters are loaded on the nanocomposites under the same load, the cracks formed in the nanocomposite produced with the high SPS applied pressure (80 MPa) probably occur because the nanocomposite has lower ductility than the nanocomposite produced with the low SPS applied pressure (40 MPa). Therefore, we conclude that the nanocomposite obtained with the high SPS applied pressure cracks at lower bending load than that obtained with the low SPS applied pressure, which leads to more rapid fracture of the nanocomposite. As a result, the bending strength decreases with increasing SPS applied pressure.

Fig. 12

SEM images of the indentations produced by Vickers hardness testing of 20 vol% TiCp/Mg nanocomposites fabricated with SPS applied pressures of 40 MPa (a) and (b), and 80 MPa (c) and (d) (SPS holding time 3.6 ks).

4. Conclusions

We have fabricated 20 vol% TiCp/Mg nanocomposites by MA, MM, and SPS with different sintering conditions to evaluate the effects of the SPS applied pressure and holding time on the mechanical properties. Based on the results, the following conclusions can be drawn:

  • (1)   From the results of XRD, TEM observation, and STEM-EDS analysis, different SPS conditions appear to have little effect on the microstructure of the TiCp/Mg nanocomposite
  • (2)   From the results of TEM–EDS, the microstructure contains many ultrafine TiC and TiC/C particles with average particle sizes of approximately 9 nm, and these particles aggregate. MgO particles of approximately 10 nm are also present.
  • (3)   TEM observations show that the grain sizes of the crystals identified as Mg solid solutions are 8–11 nm regardless of the SPS processing conditions.
  • (4)   As the applied pressure increases, the density only slightly changes for a SPS holding time of 0.6 ks but it increases for a SPS holding time of 3.6 ks.
  • (5)   Regardless of the SPS holding time, the Vickers hardness increases and the bending strength decreases with increasing applied pressure, mainly because of an increase in the compressive residual stress within the nanocomposite.

REFERENCES
 
© 2017 The Japan Institute of Metals and Materials
feedback
Top