TiC–Al2O3 composites synthesized by reactive hot pressing

TiC–Al2O3 composites synthesized by reactive hot pressing

Materials Science & Engineering A 571 (2013) 137–143 Contents lists available at SciVerse ScienceDirect Materials Science & Engineering A journal ho...

2MB Sizes 0 Downloads 56 Views

Materials Science & Engineering A 571 (2013) 137–143

Contents lists available at SciVerse ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Microstructures and mechanical properties of Ti3SiC2/TiC–Al2O3 composites synthesized by reactive hot pressing Yanzhi Cai n, Hongfeng Yin, Liqing Pan, Panjun Chen, Gaolei Sun College of Materials and Mineral Resources, Xi’an University of Architecture and Technology, 13#, Yanta Road, Xi’an, Shaanxi 710055, PR China

a r t i c l e i n f o

abstract

Article history: Received 28 October 2012 Received in revised form 5 February 2013 Accepted 8 February 2013 Available online 16 February 2013

Ti3SiC2/TiC–Al2O3 composites with different Al2O3 contents were fabricated by in-situ reaction and hot pressing sintering. Laminar Ti3SiC2 grains and granular Al2O3 grains were densely packed and tightly bonded, and cubic TiC grains presented in the surfaces of Ti3SiC2 grains. Al2O3 significantly restrained the grain growth of Ti3SiC2 matrix. The dispersed Al2O3 grains inclined to be pulled out, but Al2O3 aggregates inclined to be cut by the crack front at the crack surface. The flexural strength and fracture toughness first increased and then decreased with the increasing Al2O3 content. The composite with 20 wt% Al2O3 addition showed the highest flexural strength of 649 MPa and with 10 wt% Al2O3 addition showed the best fracture toughness of 7.15 MPa m1/2. The mechanism responsible for improved mechanical properties for Ti3SiC2/TiC–Al2O3 composites were the synergistic action of particulate dispersion reinforcement, fine-grain toughening, grain pullout, microcrack deflection, and lamella bending and slipping from three different kinds of grains. & 2013 Elsevier B.V. All rights reserved.

Keywords: Mechanical properties Ti3SiC2/TiC–Al2O3 composites In-situ reaction Microstructures

1. Introduction Ti3SiC2 with lamellar structure is just like BN or graphite. However, it is stronger in mechanical properties and better in oxidation resistance than BN or graphite [1]. It is considered a potential structural/functional material for its combined metallicand ceramic-like properties, such as low density, high modulus, good thermal and electrical conductivity, excellent thermal shock resistance and high damage tolerance and easy machinability [2–4]. However, the relatively low strength and hardness limit its application. Al2O3 has high hardness, melting point, mechanics strength and elastic modulus, and excellent adaptability in oxidation atmosphere at high-temperature. Al2O3 is a ceramic showing considerable promise for use in a number of engineering applications. TiC has high modulus, high melting point, high hardness and good erosion resistance. In fact, TiC phase usually coexists and shows special orientation relationship with Ti3SiC2 phase during the synthesis process of bulk Ti3SiC2 [4]. However, the potential of the ceramic materials has been limited by low toughness. Improving the fracture resistance of Al2O3 or TiC ceramic via microstructural design by introducing the second phase is a promise way. Hard ceramic particles such as TiC [5–7], TiB2 [5,8] and SiC [1,8,9] have been incorporated into Ti3SiC2 to improve the

n

Corresponding author. Tel./fax: þ 86 29 82205245. E-mail address: [email protected] (Y. Cai).

0921-5093/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2013.02.017

mechanical properties. Wang et al. [10,11] reported the increase by 50% in the hardness for Ti3SiC2/20 vol%Al2O3 composite but decrease in the other mechanical properties with Al2O3 addition higher than 5–10 vol%. Luo et al. [12,13] prepared Al2O3–Ti3SiC2 composites and its functionally graded materials, showing the hardness decreased but fracture toughness and strength increased with the increase of Ti3SiC2 content. Chin et al. [14] improved the strength and toughness of Al2O3 by adding Ti3SiC2 particles into Al2O3 matrix. Chen et al. [15] synthesized Ti3AlC2/TiC–Al2O3 composite in a 3TiO2–5Al–2C system which showed higher flexural strength and Vickers hardness than pure Ti3AlC2 ceramic. By a synergy mechanism between two or more strengthening and toughening methods, the enhancement in mechanical properties can be brought by the combination of Ti3SiC2 with Al2O3 and TiC. Most importantly, compared to SiC, Al2O3 and TiC are more suitable candidate reinforcements for the Ti3SiC2 matrix due to the better thermal expansion match. The thermal expansion coefficients of both Al2O3 and TiC are lower than that of Ti3SiC2 but higher than that of SiC. Additionally, compared to TiB2, Al2O3 has lower density and better oxidation resistance at high temperatures. However, previous studies rarely concentrated on the effects of the combination of Al2O3 and TiC on the microstructure and mechanical properties of Ti3SiC2-based composites. The present work focused on the microstructures and mechanical properties of Ti3SiC2/TiC–Al2O3 composites with different Al2O3 added amount from 5 wt% to 30 wt% fabricated by in situ reactive/hot pressing sintering. Ti3SiC2 matrix was synergistically reinforced by granular Al2O3 and equiaxed TiC grains in these composites.

138

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

Too much content of reinforcements resulting in the inhomogeneous phase distribution and adversely affecting the mechanical properties of the composites was indicated in previous reported papers. However, different fracture modes between dispersed particles and particle agglomeration as reinforcements were not elaborated in detail. In this study, the detailed comparisons of different fracture modes between dispersed particles and particle agglomeration as well as between large particles and fine particles were drawn. The mechanism responsible for improved mechanical properties of Ti3SiC2/TiC–Al2O3 composites from three different grains was revealed.

2. Experimental procedures 2.1. Sample preparation Ti3SiC2-based composites were prepared by in situ reaction combined with hot-pressure sintering. The Al2O3 powder was added to improve the mechanical properties of Ti3SiC2-based composites. The addition amount of Al2O3 powder was 5–30 wt% with an interval of 5 wt%. For ease of reference, the composite samples with the different Al2O3 powder addition amount would be named TA5, TA10, TA15, TA20, TA25 and TA30 respectively. The sample without Al2O3 addition named as T was also synthesized for comparison.

The starting mixtures of Ti (average particle size: 40.0 mm, 499.5% purity), Si (average particle size: 40.0 mm, 499.5% purity), carbon (average particle size: 6.27 mm, 499.5% purity) in molar ratio of 3:1:2 combined with Al2O3 (average particle size: 6.23 mm, 499.5% purity) at different mass contents were prepared by wet ball milling in ethanol for 6 h and dried. Zhang et al. [16] reported Ti3SiC2 was formed through the reaction between Ti5Si3Cx, TiCx and carbon mainly at 1400–1500 1C based on the starting materials of Ti, Si and graphite by hot pressing sintering. Song et al. [8] reported the decomposition of Ti3SiC2 at 1550–1600 1C in reactive hotpressed (TiB2 þ SiC)/Ti3SiC2 composites. Therefore 1500 1C was chosen as the sintering temperature in this experiment. The final mixture with the desired composition was then reactive hotpressed under a pressure of 25 MPa in a graphite die coated with BN under a vacuum atmosphere at 1500 1C for 3 h to obtain a dense composite sample. A heating rate of 10 1C/min was used to heat up to 1500 1C following with cooling to room temperature in the furnace. The samples with cylinders of 50 mm in diameter and about 10 mm in height were obtained.

2.2. Characterization The open porosities and bulk densities were measured by Archimedes’ method according to ASTM C-20 standard. The phase compositions and relative content of several phases in the

Fig. 1. Microscopic morphology of sintered samples: (A) TA10; (b) TA20; (c) TA30 and (d) etched SEM microstructure of TA15.

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

composites were determined by X-ray diffraction (XRD, Rigaku D/ max-2400, Japan) using powders drilled from the bulk samples. The XRD data were collected by a step-scanning diffractometer with Cu Ka radiation. The data used for quantitative analysis had an accuracy better than 0.021. The mass fraction of a phase was calculated using the equation: I =K W i ¼ Pn i i  100% i ¼ 1 Ii =K i

139

TA30

TA25 TA20

ð1Þ

TA15

where Ii is the intensity of selected lines in the diffraction pattern of i phase, Wi the mass fraction of i phase, Ki the reference intensity of i phase. The indention morphology was made using a load of 49 N with a dwell time of 15 s. The microstructural features of the composites were examined using scanning electron microscopy (SEM, VEGA3 TESCAN) equipped with an energydispersive spectroscopy (EDS) system. To expose the Ti3SiC2, Al2O3 and TiC grains, samples were mechanically polished and etched by an HNO3:HF:H2O (1:1:2) solution before SEM observation.

TA10 TA5 T

10

20

30

40

50

60

70

80

Fig. 2. XRD pattern of sintered samples.

2.3. Mechanical behavior The mechanical behavior was studied using the three-point flexural strength test (SANS CMT 4304, Sans Materials Testing Co., Shenzhen, China) and fracture toughness test. One group of samples with size of 3  4  40 mm3 was used for flexure strength measurement, with a 30 mm span at 0.5 mm/min crosshead speed. Another group of samples with size of 3  4  30 mm3 was used for fracture toughness measurement by the single edge precracked beam method using the three-point bending method with a 20 mm span at 0.05 mm/min crosshead speed. The precrack was 0.2 mm in width and 2 mm in depth along the thickness direction of samples. The flexure strength was calculated as

sf ¼

3PL 2

2bh

Table 1 Phase compositions of sintered samples/ wt%. Compositions

T

TA5

TA10

TA15

TA20

TA25

TA30

Ti3SiC2 TiC Al2O3

82.8 17.2 —

79.0 16.3 4.7

75.4 14.5 10.1

72.5 14.6 12.9

68.2 14.3 17.5

65.2 12.7 22.1

61.1 12.3 26.6

ð2Þ

where sf is flexural strength, P the maximum load at the rupture of a sample, L span, b width and h thickness of a sample. The fracture toughness was calculated as pffiffiffi   a  a 2  a 3  a 4  3PL a K IC ¼ 1:933:07 25:07 þ 25:80 þ 14:53 2 h h h h 2bh ð3Þ where KIC is the fracture toughness factor, P the maximum load at the rupture of a sample, L span, a depth of the precrack, b width and h thickness of a sample. Five samples were tested to obtain the average value for a kind of sample. A continuous record of load–displacement curve by the tester, selecting the maximum load P and testing the thickness h and width b of a sample, the flexure strength sf or fracture toughness K IC then could be calculated based on Eqs. (2) and (3).

3. Results and discussion 3.1. Microstructure characterization SEM backscattered electron image observation results of the synthesized Ti3SiC2/TiC–Al2O3 composites are shown in Fig. 1. The microstructure of composites consisted of Al2O3 grains (black), Ti3SiC2 grains (grey) and TiC grains. While TiC grains were not obvious for the average atomic weight of TiC is approximinately equal to that of Ti3SiC2. A uniform dispersion of Al2O3 phase in Ti3SiC2 matrix was clearly evident for TA10 and TA20, and there were a few small aggregates with the maximum size not

Fig. 3. Densities and open porosities of sintered samples.

exceeding 20 mm (Fig. 1(a) and (b)). But with the increase of Al2O3 addition, the dispersion of Al2O3 phase in Ti3SiC2 matrix became worse. The agglomeration of Al2O3 became significant and the sizes of agglomerations increased in TA30 (Fig. 1(c)). The grain larger than 8 mm could be considered an agglomerate which was partially sintered together due to the average grain size of starting Al2O3 powders was 6.23 mm. Some micropores were present in the aggregates (marked as white arrow), which prevented the tightly bonding between Al2O3 grains. Fig. 1(d) shows the microstructure of TA15 after polishing and chemically etching in a solution of HF/HNO3/H2O. The micro-laminar structure of Ti3SiC2 grains could be readily seen. The lamellar grains had sizes of about 10–15 mm in length and 4–6 mm in thickness. Granular grains of Al2O3 except the aggregates had a size of 2–8 mm

140

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

which was in agreement with the size of starting material. The grain boundary between Ti3SiC2 grains and Al2O3 grains after chemical etching shows the grains were densely packed and tightly bonded each other. Additionally, TiC grains which were equiaxed in shape

Fig. 4. Flexural strength and fracture toughness (KIC) versus Al2O3 added amount in Ti3SiC2/TiC–Al2O3 composites (error bar: standard deviation).

and less than 2 mm in size were located in the surfaces of Ti3SiC2 grains. Because TiC phase was produced as an ancillary phase during Ti3SiC2 synthesis process rather than in-situ addition, there was no TiC grain aggregate observed. XRD patterns of the synthesized Ti3SiC2/TiC–Al2O3 composites are shown in Fig. 2. The reflections of all samples were indexed as Ti3SiC2, Al2O3 and TiC phases. No other phases were detected. Ti3SiC2 was the main phase, Al2O3 and TiC were the second phases. Ti3SiC2, TiC and Al2O3 existed in their simple substances in the sintered Ti3SiC2/TiC–Al2O3 composites, and no evidence of the reaction among them was detected. Quantitative X-ray analysis results are listed in Table 1. The content of Al2O3 in the sintered composite sample was consistent with its initial addition amount. TiC was in-situ synthesized as an ancillary phase during Ti3SiC2 in-situ synthesis process. Because Al2O3 addition reduced the contents of Ti, Si and carbon in the starting materials, the contents of Ti3SiC2 and TiC in the sintered composite samples were decreased. Fig. 3 shows that the samples prepared in this work were dense (apparent porosity was not more than 0.50%). Increase of Al2O3 content had no apparent influence on the densification process of the composites. The decrease in bulk density with the increase of Al2O3 added amount was due to Al2O3 with the lower density compared with Ti3SiC2 and TiC.

Fig. 5. Micrograph of rupture surface of the Ti3SiC2-based composite: (a) T; (b) TA15; (c) TA20 and (d) TA30.

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

141

Fig. 6. SEM image of indentation morphology: (a) whole morphology for T; (b) whole morphology for TA20; (c) and (d) amplified photographs of the areas marked in (a); (e) amplified photograph of the area marked in (b) and (f) further amplified photograph of the area marked in (e).

3.2. Mechanical properties The mechanical properties of Ti3SiC2/TiC–Al2O3 composites are shown in Fig. 4. The flexural strength and fracture toughness both presented the trend of first increase and then decrease with the

increase of Al2O3 added amount. The strength of Ti3SiC2/TiC composite was 474 MPa. The strength increased to 649 MPa at 20 wt% Al2O3 added amount and the optimal value for flexural strength was obtained. All composites with Al2O3 addition were better than that without Al2O3 addition in the fracture toughness.

142

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

The fracture toughness of Ti3SiC2/TiC–Al2O3 composites reached the optimal value of 7.15 MPa  m1/2 at 10 wt% Al2O3 addition and that of Ti3SiC2/TiC composite was 5.18 MPa  m1/2. As has been demonstrated in Fig. 5, the fractured surfaces of Ti3SiC2-based composites were all rough. Ho-Duc [17] reported TiC inhibited the exaggerated grain growth of Ti3SiC2. Here the grain size of Ti3SiC2 matrix in Ti3SiC2/TiC–Al2O3 composites further reduced due to the presence of Al2O3 phase, and more the Al2O3 added amount, less the grain size of Ti3SiC2 (Fig. 5(b)– (d)). This microstructure with decreased grain size could be responsible for the increase in the flexural strength and fracture toughness. Such microstructure evolution can be postulated that the Al2O3 grains inhibited grain growth and the grain-boundary mobility. The fracture surface of Ti3SiC2/TiC composite is shown in Fig. 5(a). The fracture mode of Ti3SiC2 grains consisted of predominantly intergranular fracture, displaying the integrate lamellar microstructure and partly transgranular fracture, displaying the cleavages. The fracture mode for TiC grains was mainly intergranular fracture. The pits (marked as white arrows) and outcrops which resulted from TiC grain’s pullout were found in the fractured surface of Ti3SiC2/TiC composite. TiC grains with very fine grain size preferred being pulled out instead of fracture. As a reinforcing agent, TiC grains provided dispersion strengthening effect. The fracture surfaces of Ti3SiC2/TiC–Al2O3 composites are shown in Fig. 5(b)–(d). The pits resulted from TiC grain’s pullout were also found in the fractured surfaces of Ti3SiC2/TiC–Al2O3 composites (marked as white arrows in Fig. 5(b)). The intergranular and transgranular fractures were both significant for Al2O3 grains. The disperse Al2O3 grains inclined to be pulled out (marked as white squares in Fig. 5(b)), but Al2O3 aggregates inclined to be cut by the crack front at the crack surface (marked as white circles in Fig. 5(b) and (d)). Al2O3 has high granule strength and hardness, so dispersed small grains were unsusceptible to be sheared vertically by the crack front but left unbroken and bridged the broken surfaces, which provided dispersion strengthening and toughening effect. But Al2O3 is at the same time brittle, and the defects like micropores produced easily in the partially sintered grain aggregates (Fig. 1(c)). So the microstructure of grain aggregates was usually not as perfect as that of dispersed small grains, so the cracks would easily cross through the aggregates. Too much grain aggregates existed in the microstructure of the composite would adversely affect the fracture toughness and flexural strength of the composite. Therefore, if Al2O3 content was high enough, the fracture toughness and flexural strength would decrease. Fig. 6 is the indentation morphology for the composites without Al2O3 and with 20 wt% Al2O3 addition. The indentation morphology of T (Fig. 6(a)) shows the typical characteristic of microcrack propagation around the indentation resulting in a relative big damage area. On the contrary, for TA20 (Fig. 6(b)), localized damage occurred in the neighborhood of indentation and no microcrack propagation around the indentation was found. This phenomenon also suggests a better resistance to damage for Ti3SiC2/TiC–Al2O3 composites. This difference in the indentation morphologies was derived from the difference in microstructures of Ti3SiC2-based composites. Ti3SiC2 grains with a lamellar structure were relative large in sizes for Ti3SiC2/TiC composite and were planar in lamella surfaces, so the microcracks propagated along the grain boundary were relative long and straight (Fig. 6(a) and (c)). However, laminar Ti3SiC2 grains combined with cubic TiC grains and granular Al2O3 grains existed in Ti3SiC2/TiC–Al2O3 composites. Moreover, the grain size of Ti3SiC2 matrix in Ti3SiC2/TiC–Al2O3 composites was remarkably reduced, and Ti3SiC2 grains and Al2O3

grains were densely packed and tightly bonded each other, as mentioned above. Grain broken, grain sliding and pullout, grain bridging and pinning microcracks occurred during microcrack propagation. So the microcrack propagation path was complex and tortuous in Ti3SiC2/TiC–Al2O3 composites (Fig. 6(b), (e) and (f)). These zigzag microcrack paths consumed a lot of fracture energy, which was advantageous to the improvement of strength and toughness of the composites. Skala et al. [18] believed a microscale quasi-plasticity can be associated with grain debonding, grain sliding, diffuse microcracking, grain pullout, and grain bridging. Hence the fine-grain toughening and synergistic action from three different kinds of grains resulted in the microscale quasi-plasticity in TA20. So the damage area around the indentation in TA20 was relative small compared with that in T. For Ti3SiC2 grains, the process of lamella bending, slipping and splitting between neighboring lamellas derived from the weak bonding between lamellas would consume the fracture energy to improve the strength and toughness of the composites. This phenomenon was found in both T (Fig. 6(d)) and TA20 (Fig. 6(f)). The improved mechanical properties for Ti3SiC2/TiC–Al2O3 composites were attributed to the synergistic action of strengthening and toughening mechanisms like particulate dispersion strengthening, fine-grain toughening, grain pullout, microcrack deflection, and lamella bending and slipping from three different kinds of grains.

4. Conclusions

(1) Al2O3 addition significantly restrained the grain growth of Ti3SiC2 matrix. Laminar Ti3SiC2 grains and granular Al2O3 grains were densely packed and tightly bonded, and cubic TiC grains presented in the surfaces of Ti3SiC2 grains. (2) The dispersion of Al2O3 phase in Ti3SiC2 matrix became worse with the increasing Al2O3 added amount. The dispersed Al2O3 grains inclined to be pulled out, but Al2O3 aggregates inclined to be cut by the crack front at the crack surface. (3) As Al2O3 addition increased from 5 wt% to 30 wt%, the flexural strength and fracture toughness first increased and then decreased. The composite with 20 wt% Al2O3 addition showed the highest flexural strength of 649 MPa and with 10 wt% Al2O3 addition showed the best fracture toughness of 7.15 MPa m1/2. (4) The mechanism responsible for improved mechanical properties for Ti3SiC2/TiC–Al2O3 composites were the synergistic action of particulate dispersion strengthening, fine-grain toughening, grain pullout, microcrack deflection, and lamella bending and slipping from three different kinds of grains.

Acknowledgments The authors acknowledge the support of the Doctoral Starting up Foundation of Xi’an University of Architecture and Technology (RC1039) and the Foundation of Shaanxi scientific technology (2012k07-07) in China. References [1] [2] [3] [4] [5] [6] [7] [8]

S. Li, J. Xie, L. Zhang, L. Cheng, Mater. Lett. 57 (2003) 3048–3056. S. Li, L. Cheng, L. Zhang, Compos. Sci. Technol. 63 (2003) 813–819. F. Sato, J. Li, R. Watanable, Mater. Trans., JIM 41 (5) (2000) 605–608. M.W. Barsoum, Frog Solid St. Chem. 28 (2000) 201–281. J. Yang, L. Pan, W. Gu, T. Qiu, Y. Zhang, S. Zhu, Ceram. Int. 38 (2012) 649–655. W. Tian, Z. Sun, H. Hashimoto, Y. Du, Mater., Sci. Eng., A 526 (2009) 16–21. J. Zhang, L. Wang, W. Jiang, L. Chen, Mater. Sci. Eng., A 487 (2008) 137–143. K. Song, J. Yang, T. Qiu, L.M. Pan, Mater. Lett. 75 (2012) 16–19.

Y. Cai et al. / Materials Science & Engineering A 571 (2013) 137–143

[9] [10] [11] [12] [13] [14]

J. Zhang, L. Wang, L. Shi, W. Jiang, L. Chen, Scripta Mater. 56 (2007) 241–244. H.J. Wang, Z.H. Jin, Y. Miyamoto, Ceram. Int. 28 (2002) 931–934. H.J. Wang, Z.H. Jina, Y. Miyamoto., Ceram. Int. 29 (2003) 539–542. Y. Luo, W. Pan, S. Li, R. Wang, J. Li, Mater. Lett. 57 (2003) 2509–2514. Y. Luo, S. Li, J. Chen, R. Wang, J. Li, W. Pan, Mater. Res. Bull. 38 (2003) 69–78. Y.–L. Chin, W.–H. Tuan, J.–L. Huang, C.–A. Wan, J. Alloys Compd. 491 (2010) 477–482.

143

[15] J.X. Chen, J.L. Li, Y.C. Zhou, J. Mater. Sci. Technol. 22 (4) (2006) 455–458. [16] H.B. Zhang, Y.C. Zhou, Y.W. Bao, M.S. Li, J.Y. Wang, J. Eur. Ceram. Soc. 26 (2006) 2373–2380. [17] L.H. Ho-Duc, A magisterial thesis of Drexel University 2002. [18] R.D. Skala, P. Manurung, I.M. Low, Composites Part B 37 (2006) 466–480.