- Email: [email protected]
Synthesis and characterization of TiB2-reinforced iron-based composites
Journal of Materials Processing Technology 172 (2006) 70–76
Synthesis and characterization of TiB2-reinforced iron-based composites Animesh Anal, T.K. Bandyopadhyay, Karabi Das ∗ Department of Metallurgical and Materials Engineering, Indian Institute of Technology, Kharagpur 721302, India Received 22 July 2005; accepted 12 September 2005
Abstract The TiB2 -reinforced iron matrix composite (Fe-TiB2 ) was synthesized by a simple, cost-effective process involving aluminothermic reduction of blue dust (mainly Fe2 O3 ), titanium dioxide (TiO2 ) and boron trioxide (B2 O3 ) powder. Aluminothermic reduction of these oxides, being highly exothermic in nature, essentially leads to a self-propagating high-temperature synthesis (SHS) of TiB2 -reinforced Fe-based composite. The composite has been subsequently characterized by scanning electron microscopy (SEM), image analysis, X-ray diffraction (XRD), and hardness measurement. It has been found that along with TiB2 , Fe2 B also forms during the synthesis of composite. Composite, synthesized by this process, possesses high hardness and high temperature stability. The abrasive wear resistance of the composite has been compared with that of a standard wear resistant material, i.e., high-chromium white cast iron, and found to be better than that of the standard material. © 2005 Elsevier B.V. All rights reserved. Keywords: Iron matrix composite; TiB2 ; Aluminothermic reduction; Microstructure; Hardness; Wear resistance
1. Introduction Search for superior wear resistant materials has been allotted a high priority in the field of materials research even these days. Although a vast reserve of wear resistant materials does exist in the present age, constant research activities are being carried out to produce some kind of new materials, which are better in terms of properties and less expensive compared to the existing ones. As for example, hot work tool steels are frequently used in die-castings, forging dies, punches and several other parts for hot working. These steels possess high machinability and toughness but suffer from inferior wear resistance. In order to increase their wear resistance, reinforcement with hard ceramic particles could be beneficial [1]. Apart from this specific case, composite materials with steel matrix and ceramic particle reinforcements provide a scope of producing relatively inexpensive wear resistant parts. Recently, however a limited numbers of iron and steel-based composites have emerged, which are inexpensive, versatile, and exhibit relatively good mechanical properties [2]. The most commonly used ceramic particles for reinforcement of various types of steel matrices include various oxides (e.g., Al2 O3 and ZrO2 ), nitrides (e.g., TiN and Si3 N4 ), and carbides (e.g., TiC, Cr3 C2 , VC, and B4 C) [3]. A majority of these
∗
Corresponding author. Tel.: +91 3222 283254; fax: +91 3222 25503. E-mail address: [email protected] (K. Das).
0924-0136/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2005.09.011
composite materials focuses on using TiC as the reinforcing particulate phase, primarily due to its capability of improving the wear resistance of the material [4–6]. Among the popular reinforcements, TiB2 is considered as the best reinforcement for the following two reasons [2]. First, it exhibits an extremely high specific modulus of 120 GPa/Mg m−3 and hence would be desirable in applications where component weight as well as stiffness is important. Second, unlike most other ceramic reinforcements, which are reactive in molten iron, TiB2 is stable in liquid Fe. Apart from these, TiB2 is also well known for its high hardness and outstanding tribological properties. It’s hardness (3400 HV) is greater than the more commonly used WC (2000 HV) and is almost as high as that of SiC (3500 HV) [7]. Furthermore, it has a high thermal conductivity (∼110 Wm−1 K−1 at 25 ◦ C) and a significantly lower coefficient of thermal expansion than steel (∼13 × 10−6 K−1 for steel and ∼7.2 × 10−6 K−1 for TiB2 ). Thus, steel-matrix composites with TiB2 as the reinforcing phase have increased stiffness, hardness, and wear resistance, along with reduced coefficient of thermal expansion and only a moderate decrease in thermal conductivity properties [2]. Powder metallurgy is the most attractive processing route for these types of particulate-reinforced metal-matrix composites. But, there are several advantages in terms of production cost and efficiency, if such composites can be processed by liquid-based routes involving the in situ formation of the filler phase [8]. In the present study, the authors have made an attempt to synthesize TiB2 -reinforced Fe-based composites by aluminothermic
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
71
reduction of Fe2 O3 , TiO2 and B2 O3 . This type of reaction involves the reduction of a metallic or a non-metallic oxide with Al to form Al2 O3 and the corresponding metal or non-metal. It is highly exothermic in nature and if it is initiated locally it can become self-sustaining. The study also includes the evaluation of the effect of heat treatment on the microstructure and abrasive wear resistance of the composite materials.
diameter was 100 mm. The experiments were carried out at different loads, e.g., 9.8, 14.7 and 19.6 N. Each test was repeated thrice. Wear rate of the specimens was computed by the weight loss technique. Prior to weighing, the specimens were cleaned with ethanol. The cumulative weight loss was converted to cumulative volume loss by dividing the weight loss by experimentally determined density. Wear rate has been calculated by using the following formula:
2. Experimental procedure
Wear data have been plotted as either cumulative volume loss or wear rate as a function of sliding distance.
Wear rate(mm3 m−1 ) =
cumulative weight loss(gm)/density(gm/mm3 ) sliding distance (m)
2.1. Materials
3. Results and discussion Blue dust, titanium dioxide, boron trioxide and aluminum powder are the main materials used for the synthesis of the composite. Blue dust is basically an oxide of iron of chemical formula Fe2 O3 with 2.5 wt.% SiO2 . Laboratory grade TiO2 (with 99% purity), B2 O3 (99.5% purity) and Al powder (with 99% purity and ∼50 m size) were used in the present study. The as-received highchromium iron contains 2.90% C, 1.00% Si, 1.00% Mn, 0.05% S, 0.032% P, 18.08% Cr, 0.80% Ni, 2.00% Mo, 0.30% Cu, and balance Fe (in wt.%) and has a hardness of Rc 60.
2.2. Composite synthesis The charge calculation was done by simple stoichiometric method considering 20 vol.% TiB2 in Fe matrix. Anticipating the loss of B2 O3 and TiO2 into the slag phase, 20 and 40% excess amounts of B2 O3 and TiO2 , respectively were added to the charge composition. Stoichiometric amount of aluminum, needed to reduce all the oxides, was added to the charge. Blue dust (375 g) was mixed with predetermined amount of TiO2 (62 g), B2 O3 (82 g) and Al powder (190 g). The charge was preheated in a pit furnace at a temperature of about 750 ◦ C for about half an hour in a zircon coated clay–graphite crucible. After preheating, the crucible was removed from the furnace and magnesium turnings were then added to trigger the self-propagating high-temperature synthesis (SHS) reaction. The heat generated due to aluminothermic reduction was high enough to melt the charge completely and uniformly. The heavier metallic phase and the lighter slag phase was well separated by gravity. A bottom pouring arrangement was made so that the liquid metal could be poured directly into a metal mould by opening the plug at the bottom of the crucible.
2.3. Microscopy, image analysis, and X-ray diffraction (XRD) study Metallographic samples of dimension 12 mm × 12 mm × 10 mm were cut from the middle portion of castings of original size. The specimens were etched with 2% nital (2 ml HNO3 + 98 ml ethanol) and were examined using an optical microscope as well as a scanning electron microscope (SEM) equipped with an energy dispersive X-ray spectrometer (EDS). Leica QWIN image analysis software was used to find out the volume fraction of different phases. X-ray diffraction (XRD) was used to analyze the presence of the phases in the composites using Co K␣ radiation.
2.4. Hardness test
3.1. Thermodynamic analysis The feasibility of the reactions involved in synthesizing TiB2 reinforced Fe-based composite has been considered. The important reactions leading to the formation of Fe-TiB2 composite can be written as: 3TiO2 (s) + 3B2 O3 (l) + 10Al(l) = 3TiB2 (s) + 5Al2 O3 (s) (1) Fe2 O3 (s) + 2Al(l) = 2Fe(l) + Al2 O3 (s)
(2)
where (s) and (1) denote solid and liquid states, respectively. It is quite likely that liquid B2 O3 reacts with liquid Al during the initial stage of SHS due to its low melting point and raises the temperature, which triggers the reduction reaction of solid TiO2 and Fe2 O3 by liquid Al and releases huge amount of heat. Iron oxide is reduced into Fe and thus forms the matrix. On the other hand reduced Ti reacts with B and forms TiB2 reinforce◦ ment. The standard free energy change, GT and the standard ◦ enthalpy change, HT for reactions (1) and (2) have been calculated using data from Kubaschewski et al. [10] and Gaskell ◦ [11]. The thermodynamic calculation indicates that GT and ◦ HT values are: For reaction (1) ◦
GT = −2 761 726 − 833 T + 160 T ln T − 87 T 2 + 27 × 105 /T J/mol ◦
HT = −2 762 027 − 160 T + 36 × 10−3 T 2 + 56 × 105 /T J/mol For reaction (2) ◦
Hardness values of the composites in as-cast as well as annealed condition were measured in Rc scale using diamond indentor and 150-kg load. The micro-hardness values of the individual phases, present in the composites, were determined using Vicker’s indentor. The average of 10 measurements has been taken as the hardness of the material/individual phase.
GT = −834 399 + 223 T − 31 T ln T + 35 × 10−3 T 2 − 5 × 105 /T J/mol ◦
HT = −854 398 + 30 T − 35 × 10−3 T 2 − 12 × 105 /T J/mol
2.5. Abrasive wear test ◦
Abrasive wear tests were carried out on 12 mm × 12 mm × 10 mm samples of composite and high-Cr iron, against 220 grit SiC paper affixed to a rotating flat disc of 250 mm diameter [9]. The rotating speed was 500 r.p.m. and the duration of the test was 12 min. The sliding velocity was fixed at 2.61 ms−1 and track
◦
The GT and HT for both reactions (1) and (2) are plotted in Fig. 1 as a function of temperature. For both reactions they are negative and hence, these reactions can occur with evolution of heat. The first and foremost requirement for a reaction
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
72
Fig. 3. X-ray diffraction pattern of as-cast composite using Co K␣ radiation.
to become self-sustaining is that the reaction must be highly exothermic. It is observed from Fig. 1 that the reactions are highly exothermic in nature, indicating that the reactions will be self-sustaining. 3.2. As-cast microstructure
◦
◦
Fig. 1. Variation of GT and HT as a function of temperature for the reactions (a) 3TiO2 (s) + 3B2 O3 (l) + 10 Al (l) = 3TiB2 (s) + 5Al2 O3 (s) and (b) Fe2 O3 (s) + 2Al (l) = 2Fe(l) + Al2 O3 (s).
As-cast microstructure of composite shows the presence of interdendritic eutectic-type phase together with particulates dispersed throughout the iron matrix (Fig. 2). The XRD of the composite is shown in Fig. 3. The particulates and the interdendritic eutectic phase have been identified as TiB2 and Fe2 B, respectively. The volume fractions of TiB2 and Fe2 B, as determined by image analysis, are 8 and 35 vol.%, respectively. Iron, Ti, and B get reduced from their respective oxides. Ti reacts with the B to form the TiB2 particles. Thus, the
Fig. 2. Scanning electron micrographs of as-cast composite showing (A) the distribution of TiB2 particles in Fe matrix, (B) both large (l) and small (s) particles of TiB2 , (C) eutectic (e) structure, and (D) sympathetic nucleation of TiB2 particles.
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
reinforcing phase in the composites is generated in situ. It is quite likely that a supersaturated solution of Ti and B in liquid Fe forms with simultaneous formation of TiB2 through the reaction Ti + B(s) = TiB2 (s), where Ti means Ti in solution. During cooling of the melt the solubility of both Ti and B in Fe decreases [12,13]. Therefore, excess amounts of Ti and B forms TiB2 particles mainly by Ti + B = TiB2 (s) reaction, i.e., solution-precipitation mechanism. The TiB2 particles that formed at high temperature by the reaction Ti + B(s) = TiB2 (s) had enough time to grow whereas the TiB2 particles formed by the reaction Ti + B = TiB2 (s) did not have enough time to grow. As a result, distribution of both large and small particles of TiB2 can be seen in the as-cast microstructure (Fig. 2). One interesting thing to note is the morphology of TiB2 particles. Some TiB2 particles appear to be sympathetically nucleated at the already existing TiB2 precipitates. Sympathetic nucleation can be defined as the nucleation of a precipitate at the interphase boundary of another precipitate of the same phase. Replacement of a high-energy precipitate-matrix interface by a low energy precipitate-precipitate interface decreases the activation energy barrier for sympathetic nucleation [14]. Iron along with Ti, B and Si (Si comes from the reduction of SiO2 , which is present in the blue dust) forms a liquid solution. The whole amount of Ti that is present in the solution gets con-
73
Fig. 4. EDS spectra taken from the matrix.
sumed during the formation of TiB2 . As a result no Ti is present in the matrix. This is evident from the EDS spectra taken from the matrix (Fig. 4). Although 40% extra of the required stoichiometric amount of TiO2 was added, it appears that some unreacted
Fig. 5. Scanning electron micrographs of composite annealed at 900 ◦ C for (A and B) 3 h, (C and D) 6 h, and (E and F) 9 h.
74
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76 Table 2 Microhardness of different phases of composite
Fig. 6. Line-scan taken across the TiB2 particles showing no diffusion of Ti from particles to matrix after annealing at 900 ◦ C for 9 h.
TiO2 joined the slag phase. As a result some B remained in the solution of Fe, after all the Ti was consumed during the formation of TiB2 . The excess B that remained in the solution of Fe precipitated as Fe2 B as the solubility of B in Fe is virtually zero at room temperature. 3.3. Annealed microstructure In-order to assess the microstructural stabilities of the as-cast composite, it was subjected to annealing at 900 ◦ C for different times (3, 6, and 9 h). The SEM micrographs of the composite, annealed at 900 ◦ C for different times, are shown in Fig. 5. It is observed that the microstructures have not changed at all due to annealing. This shows that synthesized composite is stable at high temperature. Fig. 6 shows the line scan taken across the TiB2 particles from the composite which was annealed at 900 ◦ C for 9 h. It is observed that there is no diffusion of Ti has occurred as a result of annealing which suggests that particles are stable at high temperature.
Material
Matrix (HV 50 g)
Fe2 B (HV 50 g)
TiB2 (HV 100 g)
As-cast composite Annealed composite (900 ◦ C, 9 h)
383 380
831 833
3261 3259
hardness value of TiB2 vary from 2900 HV [16] to 3400 HV [7]. The measured hardness value of TiB2 falls within the reported range. The measured hardness of Fe2 B matches with the literature value [2]. There is no significant change in hardness values of the composite after annealing at 900 ◦ C for 3, 6 and 9 h. This is due to the fact that there are no changes in microstructure and also in the microhardness of the matrix, Fe2 B and TiB2 particle of the composite due to annealing treatment. Therefore, these composites can be expected to be stable at high temperature. 3.5. Abrasive wear The main purpose of the production of the iron matrix composite is for the applications like feed screws, underwater pelletizer knives, draw punches, draw rings, pill dies, powder compaction dies, mill guide rolls, work rolls, etc. requiring exceptional wear resistance [17]. Hence, abrasive wear behavior of the synthesized composite has been evaluated and compared with a high-chromium iron. The cumulative volume loss and wear rate of the composite and that of the high-chromium iron at different loads as a function of sliding distance are shown in Figs. 7 and 8, respectively. Fig. 7 shows that material loss is more for high-chromium iron than that for the as-cast composite at all loads. Although, the hardness of the as-cast material is lower than the reference material, the abrasive wear resistance property of the as-cast composite is better than the reference material. A composite material’s wear resistance property is not dictated by hardness alone. It also depends on microstructural parameters like volume fraction,
3.4. Hardness The hardness values of the as-cast and annealed composite are presented in Table 1. The microhardness values of the different phases in the as-cast condition and after annealing at 900 ◦ C for 9 h are shown in Table 2. The hardness of the matrix is significantly higher than that of plain carbon steel matrix. This is due to solid solution strengthening of iron by aluminum and silicon. The reported hardness value of plain carbon steel with 0.2% C in the annealed condition is 150 HV [15]. The reported Table 1 Hardness of the Fe-TiB2 composite Material
Hardness (Rc )
As-cast Annealed at 900 ◦ C for 3 h Annealed at 900 ◦ C for 6 h Annealed at 900 ◦ C for 9 h
52 52 51 51
Fig. 7. Wear volume of as-cast composite and high-chromium iron at various loads.
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
Fig. 8. Wear rate of as-cast composite and high-chromium iron at various loads.
size, shape, and distribution of embedded particles, properties of matrix, and the interfacial bonding between the two phases [18]. It is evident from Fig. 8 that wear rate is more at the initial stage and then decreases with the increase in sliding distance at all loads. Initially, the relatively soft matrix controls the wear rate. However, at large sliding distance, hard particles exposed to the surface, bear the load resulting in a decrease in the wear rate. Better wear resistance of the as-cast composite compared to high-chromium iron may be attributed to the fact that highchromium iron contains M7 C3 type carbide having hardness between 1200 and 1500 HV [19]. On the other hand, the hardness of TiB2 particles in as-cast composite is 3260 HV. Khrushov et al. [20] suggests that higher is the hardness of the reinforcing particle, better is the wear resistance. Both TiB2 (8 vol.%) and Fe2 B (35 vol.%) phases present in the composite may act as reinforcements in the composite. The amount of carbide in high-chromium iron, as determined by image analysis, is about 30 vol.%. Hence, the composite contains higher volume percent of reinforcement compared to volume fraction of M7 C3 carbide in high chromium iron. It is evident from the SEM micrograph of the worn surface of the composite (Fig. 9) that no particle
75
Fig. 10. Wear volume of as-cast composite, annealed composite (annealed at 900 ◦ C for 9 h) and high-chromium iron at a load of 9.8 and 19.6 N.
pull-out has taken place even at an applied load of 19.6 N. It indicates that TiB2 , having very high specific modulus, supports the load bearing capacity and there is a good bonding between the reinforcing TiB2 particles and the iron matrix. The wear behaviour of as-cast composite, annealed composite (annealed at 900 ◦ C for 9 h) and high-chromium iron at loads of 9.8 and 19.6 N is shown in Fig. 10. It is observed from the figure that there is no significant change in the wear behaviour of as-cast and annealed composites. This is due to the fact that there are no changes in the microstructure, bulk hardness and microhardness of the individual phases of as-cast composite after annealing. 4. Conclusions It is possible to synthesize Fe-TiB2 composite by aluminothermic reduction of Fe2 O3 , TiO2 and B2 O3 . The aluminothermic reduction of Fe2 O3 , TiO2 and B2 O3 , being highly exothermic in nature, essentially leads to a self-propagating high-temperature synthesis of the Fe-TiB2 composite. This is verified both thermodynamically and experimentally. The formation of Fe2 B, along with TiB2 , could not be avoided during the synthesis of the composite. However, it can possibly be avoided by increasing the amount of Ti or by decreasing the amount of B in the melt. The abrasive wear resistance of the composite is better than that of a high-chromium iron. The composite also possesses good high temperature stability. TiB2 particles do not undergo dissolution or coarsening during annealing at 900 ◦ C. References
Fig. 9. SEM micrograph of the worn surface of as-cast composite at a load of 19.6 N.
[1] E. Pagounis, M. Talvitie, V.K. Lindross, Influence of matrix structure on the abrasion wear resistance and toughness of a hot isostatic pressed white iron matrix composite, Metall. Mater. Trans. A 27 (1997) 4183–4191. [2] C.C. Degnan, P.H. Shipway, The incorporation of self-propagating, hightemperature synthesis-formed Fe-TiB2 into ferrous melt, Metall. Mater. Trans. A 33 (2002) 2973–2983. [3] E. Pagounis, M. Talvitie, V.K. Lindross, Influence of reinforcement volume fraction and size on the microstructure and abrasion wear resistance
76
[4]
[5]
[6]
[7] [8]
[9]
[10]
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76 of hot isostatic pressed white iron matrix composite, Metall. Mater. Trans. A 27 (1997) 4171–4181. T.Z. Kattamis, T. Sugamuma, Solidification processing and tribological behavior of particulate TiC-ferrous matrix composites, Mater. Sci. Eng. A 128 (1990) 241–252. S. Skolianos, T.Z. Kattamis, M. Chen, B.V. Chambers, Cast microstructure and tribological properties of particulate TiC-reinforced Ni-base or stainless steel matrix composites, Mater. Sci. Eng. A 183 (1994) 195–204. C.C. Degnan, P.H. Shipway, J.V. Wood, Elevated temperature sliding wear behaviour of TiC-reinforced steel matrix composites, Wear 251 (2001) 1444–1451. B.S. Terry, O.S. Chinyamakobvu, Dispersion and reaction of TiB2 in liquid iron alloy, Mater. Sci. Technol. 8 (1992) 491–499. B.S. Terry, O.S. Chinyamakobvu, Carbothermic reduction of ilmenite and rutile as means of production of iron based Ti(O,C) metal matrix composites, Mater. Sci. Technol. 7 (1991) 842–848. P.P. Bandyopadhyay, Processing and characterization of plasma sprayed ceramic coatings on steel substrate, PhD thesis, Indian Institute of Technology-Kharagpur, India, 2001. O. Kubaschewski, C.B. Alcock, P.J. Spencer, Materials Thermochemistry, Pergamon press, Oxford, 1993, p. 258.
[11] D.R. Gaskell, Introduction to Metallurgical Thermodynamics, second ed., McGraw-Hill Book Company, New York, USA, 1981, pp. 585– 587. [12] T.B. Massalski, Binary Alloy Phase Diagrams, 1, Second ed., ASM International, Materials Park, OH 44073, 2001, p. 482. [13] T.B. Massalski, Binary Alloy Phase Diagrams, 2, Second ed., ASM International, Materials Park, OH 44073, 2001, p. 1785. [14] H.I. Aaronson, G. Spanos, R.A. Masamura, R.G. Vardiman, D.W. Moon, E.S.K. Menon, M.J. Hall, Sympathetic nucleation: an overview, Mater. Sci. Eng. B 32 (1995) 107–123. [15] H.E. Boyer, T.L. Gall (Eds.), Metals Handbook, desk ed., American Society for Metals, Metals Park, OH, 1985, p. 420. [16] H. Holleck, H. Leiste, W. Schneider, Significance of phase boundaries in wear resistant TiC/TiB2 materials, R an HM (1987) 149–154. [17] V.K. Rai, R. Srivastava, S.K. Nath, S. Ray, Wear in cast titanium carbide reinforced ferrous composites under dry sliding wear, Wear 231 (1991) 265–271. [18] E. Pagounis, M. Talvitie, V.K. Lindroos, Microstructure and mechanical properties of hot work tool steel matrix composites produced by hot isostatic pressing, Powder Metall. 40 (1997) 55–61. [19] Y. Geller, Tool Steels, Mir Publishers, Moscow, 1978, p. 186. [20] M.M. Khruschov, Principles of abrasive wear, Wear 69 (1974) 69–88.
Synthesis and characterization of TiB2-reinforced iron-based composites Animesh Anal, T.K. Bandyopadhyay, Karabi Das ∗ Department of Metallurgical and Materials Engineering, Indian Institute of Technology, Kharagpur 721302, India Received 22 July 2005; accepted 12 September 2005
Abstract The TiB2 -reinforced iron matrix composite (Fe-TiB2 ) was synthesized by a simple, cost-effective process involving aluminothermic reduction of blue dust (mainly Fe2 O3 ), titanium dioxide (TiO2 ) and boron trioxide (B2 O3 ) powder. Aluminothermic reduction of these oxides, being highly exothermic in nature, essentially leads to a self-propagating high-temperature synthesis (SHS) of TiB2 -reinforced Fe-based composite. The composite has been subsequently characterized by scanning electron microscopy (SEM), image analysis, X-ray diffraction (XRD), and hardness measurement. It has been found that along with TiB2 , Fe2 B also forms during the synthesis of composite. Composite, synthesized by this process, possesses high hardness and high temperature stability. The abrasive wear resistance of the composite has been compared with that of a standard wear resistant material, i.e., high-chromium white cast iron, and found to be better than that of the standard material. © 2005 Elsevier B.V. All rights reserved. Keywords: Iron matrix composite; TiB2 ; Aluminothermic reduction; Microstructure; Hardness; Wear resistance
1. Introduction Search for superior wear resistant materials has been allotted a high priority in the field of materials research even these days. Although a vast reserve of wear resistant materials does exist in the present age, constant research activities are being carried out to produce some kind of new materials, which are better in terms of properties and less expensive compared to the existing ones. As for example, hot work tool steels are frequently used in die-castings, forging dies, punches and several other parts for hot working. These steels possess high machinability and toughness but suffer from inferior wear resistance. In order to increase their wear resistance, reinforcement with hard ceramic particles could be beneficial [1]. Apart from this specific case, composite materials with steel matrix and ceramic particle reinforcements provide a scope of producing relatively inexpensive wear resistant parts. Recently, however a limited numbers of iron and steel-based composites have emerged, which are inexpensive, versatile, and exhibit relatively good mechanical properties [2]. The most commonly used ceramic particles for reinforcement of various types of steel matrices include various oxides (e.g., Al2 O3 and ZrO2 ), nitrides (e.g., TiN and Si3 N4 ), and carbides (e.g., TiC, Cr3 C2 , VC, and B4 C) [3]. A majority of these
∗
Corresponding author. Tel.: +91 3222 283254; fax: +91 3222 25503. E-mail address: [email protected] (K. Das).
0924-0136/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2005.09.011
composite materials focuses on using TiC as the reinforcing particulate phase, primarily due to its capability of improving the wear resistance of the material [4–6]. Among the popular reinforcements, TiB2 is considered as the best reinforcement for the following two reasons [2]. First, it exhibits an extremely high specific modulus of 120 GPa/Mg m−3 and hence would be desirable in applications where component weight as well as stiffness is important. Second, unlike most other ceramic reinforcements, which are reactive in molten iron, TiB2 is stable in liquid Fe. Apart from these, TiB2 is also well known for its high hardness and outstanding tribological properties. It’s hardness (3400 HV) is greater than the more commonly used WC (2000 HV) and is almost as high as that of SiC (3500 HV) [7]. Furthermore, it has a high thermal conductivity (∼110 Wm−1 K−1 at 25 ◦ C) and a significantly lower coefficient of thermal expansion than steel (∼13 × 10−6 K−1 for steel and ∼7.2 × 10−6 K−1 for TiB2 ). Thus, steel-matrix composites with TiB2 as the reinforcing phase have increased stiffness, hardness, and wear resistance, along with reduced coefficient of thermal expansion and only a moderate decrease in thermal conductivity properties [2]. Powder metallurgy is the most attractive processing route for these types of particulate-reinforced metal-matrix composites. But, there are several advantages in terms of production cost and efficiency, if such composites can be processed by liquid-based routes involving the in situ formation of the filler phase [8]. In the present study, the authors have made an attempt to synthesize TiB2 -reinforced Fe-based composites by aluminothermic
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
71
reduction of Fe2 O3 , TiO2 and B2 O3 . This type of reaction involves the reduction of a metallic or a non-metallic oxide with Al to form Al2 O3 and the corresponding metal or non-metal. It is highly exothermic in nature and if it is initiated locally it can become self-sustaining. The study also includes the evaluation of the effect of heat treatment on the microstructure and abrasive wear resistance of the composite materials.
diameter was 100 mm. The experiments were carried out at different loads, e.g., 9.8, 14.7 and 19.6 N. Each test was repeated thrice. Wear rate of the specimens was computed by the weight loss technique. Prior to weighing, the specimens were cleaned with ethanol. The cumulative weight loss was converted to cumulative volume loss by dividing the weight loss by experimentally determined density. Wear rate has been calculated by using the following formula:
2. Experimental procedure
Wear data have been plotted as either cumulative volume loss or wear rate as a function of sliding distance.
Wear rate(mm3 m−1 ) =
cumulative weight loss(gm)/density(gm/mm3 ) sliding distance (m)
2.1. Materials
3. Results and discussion Blue dust, titanium dioxide, boron trioxide and aluminum powder are the main materials used for the synthesis of the composite. Blue dust is basically an oxide of iron of chemical formula Fe2 O3 with 2.5 wt.% SiO2 . Laboratory grade TiO2 (with 99% purity), B2 O3 (99.5% purity) and Al powder (with 99% purity and ∼50 m size) were used in the present study. The as-received highchromium iron contains 2.90% C, 1.00% Si, 1.00% Mn, 0.05% S, 0.032% P, 18.08% Cr, 0.80% Ni, 2.00% Mo, 0.30% Cu, and balance Fe (in wt.%) and has a hardness of Rc 60.
2.2. Composite synthesis The charge calculation was done by simple stoichiometric method considering 20 vol.% TiB2 in Fe matrix. Anticipating the loss of B2 O3 and TiO2 into the slag phase, 20 and 40% excess amounts of B2 O3 and TiO2 , respectively were added to the charge composition. Stoichiometric amount of aluminum, needed to reduce all the oxides, was added to the charge. Blue dust (375 g) was mixed with predetermined amount of TiO2 (62 g), B2 O3 (82 g) and Al powder (190 g). The charge was preheated in a pit furnace at a temperature of about 750 ◦ C for about half an hour in a zircon coated clay–graphite crucible. After preheating, the crucible was removed from the furnace and magnesium turnings were then added to trigger the self-propagating high-temperature synthesis (SHS) reaction. The heat generated due to aluminothermic reduction was high enough to melt the charge completely and uniformly. The heavier metallic phase and the lighter slag phase was well separated by gravity. A bottom pouring arrangement was made so that the liquid metal could be poured directly into a metal mould by opening the plug at the bottom of the crucible.
2.3. Microscopy, image analysis, and X-ray diffraction (XRD) study Metallographic samples of dimension 12 mm × 12 mm × 10 mm were cut from the middle portion of castings of original size. The specimens were etched with 2% nital (2 ml HNO3 + 98 ml ethanol) and were examined using an optical microscope as well as a scanning electron microscope (SEM) equipped with an energy dispersive X-ray spectrometer (EDS). Leica QWIN image analysis software was used to find out the volume fraction of different phases. X-ray diffraction (XRD) was used to analyze the presence of the phases in the composites using Co K␣ radiation.
2.4. Hardness test
3.1. Thermodynamic analysis The feasibility of the reactions involved in synthesizing TiB2 reinforced Fe-based composite has been considered. The important reactions leading to the formation of Fe-TiB2 composite can be written as: 3TiO2 (s) + 3B2 O3 (l) + 10Al(l) = 3TiB2 (s) + 5Al2 O3 (s) (1) Fe2 O3 (s) + 2Al(l) = 2Fe(l) + Al2 O3 (s)
(2)
where (s) and (1) denote solid and liquid states, respectively. It is quite likely that liquid B2 O3 reacts with liquid Al during the initial stage of SHS due to its low melting point and raises the temperature, which triggers the reduction reaction of solid TiO2 and Fe2 O3 by liquid Al and releases huge amount of heat. Iron oxide is reduced into Fe and thus forms the matrix. On the other hand reduced Ti reacts with B and forms TiB2 reinforce◦ ment. The standard free energy change, GT and the standard ◦ enthalpy change, HT for reactions (1) and (2) have been calculated using data from Kubaschewski et al. [10] and Gaskell ◦ [11]. The thermodynamic calculation indicates that GT and ◦ HT values are: For reaction (1) ◦
GT = −2 761 726 − 833 T + 160 T ln T − 87 T 2 + 27 × 105 /T J/mol ◦
HT = −2 762 027 − 160 T + 36 × 10−3 T 2 + 56 × 105 /T J/mol For reaction (2) ◦
Hardness values of the composites in as-cast as well as annealed condition were measured in Rc scale using diamond indentor and 150-kg load. The micro-hardness values of the individual phases, present in the composites, were determined using Vicker’s indentor. The average of 10 measurements has been taken as the hardness of the material/individual phase.
GT = −834 399 + 223 T − 31 T ln T + 35 × 10−3 T 2 − 5 × 105 /T J/mol ◦
HT = −854 398 + 30 T − 35 × 10−3 T 2 − 12 × 105 /T J/mol
2.5. Abrasive wear test ◦
Abrasive wear tests were carried out on 12 mm × 12 mm × 10 mm samples of composite and high-Cr iron, against 220 grit SiC paper affixed to a rotating flat disc of 250 mm diameter [9]. The rotating speed was 500 r.p.m. and the duration of the test was 12 min. The sliding velocity was fixed at 2.61 ms−1 and track
◦
The GT and HT for both reactions (1) and (2) are plotted in Fig. 1 as a function of temperature. For both reactions they are negative and hence, these reactions can occur with evolution of heat. The first and foremost requirement for a reaction
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
72
Fig. 3. X-ray diffraction pattern of as-cast composite using Co K␣ radiation.
to become self-sustaining is that the reaction must be highly exothermic. It is observed from Fig. 1 that the reactions are highly exothermic in nature, indicating that the reactions will be self-sustaining. 3.2. As-cast microstructure
◦
◦
Fig. 1. Variation of GT and HT as a function of temperature for the reactions (a) 3TiO2 (s) + 3B2 O3 (l) + 10 Al (l) = 3TiB2 (s) + 5Al2 O3 (s) and (b) Fe2 O3 (s) + 2Al (l) = 2Fe(l) + Al2 O3 (s).
As-cast microstructure of composite shows the presence of interdendritic eutectic-type phase together with particulates dispersed throughout the iron matrix (Fig. 2). The XRD of the composite is shown in Fig. 3. The particulates and the interdendritic eutectic phase have been identified as TiB2 and Fe2 B, respectively. The volume fractions of TiB2 and Fe2 B, as determined by image analysis, are 8 and 35 vol.%, respectively. Iron, Ti, and B get reduced from their respective oxides. Ti reacts with the B to form the TiB2 particles. Thus, the
Fig. 2. Scanning electron micrographs of as-cast composite showing (A) the distribution of TiB2 particles in Fe matrix, (B) both large (l) and small (s) particles of TiB2 , (C) eutectic (e) structure, and (D) sympathetic nucleation of TiB2 particles.
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
reinforcing phase in the composites is generated in situ. It is quite likely that a supersaturated solution of Ti and B in liquid Fe forms with simultaneous formation of TiB2 through the reaction Ti + B(s) = TiB2 (s), where Ti means Ti in solution. During cooling of the melt the solubility of both Ti and B in Fe decreases [12,13]. Therefore, excess amounts of Ti and B forms TiB2 particles mainly by Ti + B = TiB2 (s) reaction, i.e., solution-precipitation mechanism. The TiB2 particles that formed at high temperature by the reaction Ti + B(s) = TiB2 (s) had enough time to grow whereas the TiB2 particles formed by the reaction Ti + B = TiB2 (s) did not have enough time to grow. As a result, distribution of both large and small particles of TiB2 can be seen in the as-cast microstructure (Fig. 2). One interesting thing to note is the morphology of TiB2 particles. Some TiB2 particles appear to be sympathetically nucleated at the already existing TiB2 precipitates. Sympathetic nucleation can be defined as the nucleation of a precipitate at the interphase boundary of another precipitate of the same phase. Replacement of a high-energy precipitate-matrix interface by a low energy precipitate-precipitate interface decreases the activation energy barrier for sympathetic nucleation [14]. Iron along with Ti, B and Si (Si comes from the reduction of SiO2 , which is present in the blue dust) forms a liquid solution. The whole amount of Ti that is present in the solution gets con-
73
Fig. 4. EDS spectra taken from the matrix.
sumed during the formation of TiB2 . As a result no Ti is present in the matrix. This is evident from the EDS spectra taken from the matrix (Fig. 4). Although 40% extra of the required stoichiometric amount of TiO2 was added, it appears that some unreacted
Fig. 5. Scanning electron micrographs of composite annealed at 900 ◦ C for (A and B) 3 h, (C and D) 6 h, and (E and F) 9 h.
74
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76 Table 2 Microhardness of different phases of composite
Fig. 6. Line-scan taken across the TiB2 particles showing no diffusion of Ti from particles to matrix after annealing at 900 ◦ C for 9 h.
TiO2 joined the slag phase. As a result some B remained in the solution of Fe, after all the Ti was consumed during the formation of TiB2 . The excess B that remained in the solution of Fe precipitated as Fe2 B as the solubility of B in Fe is virtually zero at room temperature. 3.3. Annealed microstructure In-order to assess the microstructural stabilities of the as-cast composite, it was subjected to annealing at 900 ◦ C for different times (3, 6, and 9 h). The SEM micrographs of the composite, annealed at 900 ◦ C for different times, are shown in Fig. 5. It is observed that the microstructures have not changed at all due to annealing. This shows that synthesized composite is stable at high temperature. Fig. 6 shows the line scan taken across the TiB2 particles from the composite which was annealed at 900 ◦ C for 9 h. It is observed that there is no diffusion of Ti has occurred as a result of annealing which suggests that particles are stable at high temperature.
Material
Matrix (HV 50 g)
Fe2 B (HV 50 g)
TiB2 (HV 100 g)
As-cast composite Annealed composite (900 ◦ C, 9 h)
383 380
831 833
3261 3259
hardness value of TiB2 vary from 2900 HV [16] to 3400 HV [7]. The measured hardness value of TiB2 falls within the reported range. The measured hardness of Fe2 B matches with the literature value [2]. There is no significant change in hardness values of the composite after annealing at 900 ◦ C for 3, 6 and 9 h. This is due to the fact that there are no changes in microstructure and also in the microhardness of the matrix, Fe2 B and TiB2 particle of the composite due to annealing treatment. Therefore, these composites can be expected to be stable at high temperature. 3.5. Abrasive wear The main purpose of the production of the iron matrix composite is for the applications like feed screws, underwater pelletizer knives, draw punches, draw rings, pill dies, powder compaction dies, mill guide rolls, work rolls, etc. requiring exceptional wear resistance [17]. Hence, abrasive wear behavior of the synthesized composite has been evaluated and compared with a high-chromium iron. The cumulative volume loss and wear rate of the composite and that of the high-chromium iron at different loads as a function of sliding distance are shown in Figs. 7 and 8, respectively. Fig. 7 shows that material loss is more for high-chromium iron than that for the as-cast composite at all loads. Although, the hardness of the as-cast material is lower than the reference material, the abrasive wear resistance property of the as-cast composite is better than the reference material. A composite material’s wear resistance property is not dictated by hardness alone. It also depends on microstructural parameters like volume fraction,
3.4. Hardness The hardness values of the as-cast and annealed composite are presented in Table 1. The microhardness values of the different phases in the as-cast condition and after annealing at 900 ◦ C for 9 h are shown in Table 2. The hardness of the matrix is significantly higher than that of plain carbon steel matrix. This is due to solid solution strengthening of iron by aluminum and silicon. The reported hardness value of plain carbon steel with 0.2% C in the annealed condition is 150 HV [15]. The reported Table 1 Hardness of the Fe-TiB2 composite Material
Hardness (Rc )
As-cast Annealed at 900 ◦ C for 3 h Annealed at 900 ◦ C for 6 h Annealed at 900 ◦ C for 9 h
52 52 51 51
Fig. 7. Wear volume of as-cast composite and high-chromium iron at various loads.
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76
Fig. 8. Wear rate of as-cast composite and high-chromium iron at various loads.
size, shape, and distribution of embedded particles, properties of matrix, and the interfacial bonding between the two phases [18]. It is evident from Fig. 8 that wear rate is more at the initial stage and then decreases with the increase in sliding distance at all loads. Initially, the relatively soft matrix controls the wear rate. However, at large sliding distance, hard particles exposed to the surface, bear the load resulting in a decrease in the wear rate. Better wear resistance of the as-cast composite compared to high-chromium iron may be attributed to the fact that highchromium iron contains M7 C3 type carbide having hardness between 1200 and 1500 HV [19]. On the other hand, the hardness of TiB2 particles in as-cast composite is 3260 HV. Khrushov et al. [20] suggests that higher is the hardness of the reinforcing particle, better is the wear resistance. Both TiB2 (8 vol.%) and Fe2 B (35 vol.%) phases present in the composite may act as reinforcements in the composite. The amount of carbide in high-chromium iron, as determined by image analysis, is about 30 vol.%. Hence, the composite contains higher volume percent of reinforcement compared to volume fraction of M7 C3 carbide in high chromium iron. It is evident from the SEM micrograph of the worn surface of the composite (Fig. 9) that no particle
75
Fig. 10. Wear volume of as-cast composite, annealed composite (annealed at 900 ◦ C for 9 h) and high-chromium iron at a load of 9.8 and 19.6 N.
pull-out has taken place even at an applied load of 19.6 N. It indicates that TiB2 , having very high specific modulus, supports the load bearing capacity and there is a good bonding between the reinforcing TiB2 particles and the iron matrix. The wear behaviour of as-cast composite, annealed composite (annealed at 900 ◦ C for 9 h) and high-chromium iron at loads of 9.8 and 19.6 N is shown in Fig. 10. It is observed from the figure that there is no significant change in the wear behaviour of as-cast and annealed composites. This is due to the fact that there are no changes in the microstructure, bulk hardness and microhardness of the individual phases of as-cast composite after annealing. 4. Conclusions It is possible to synthesize Fe-TiB2 composite by aluminothermic reduction of Fe2 O3 , TiO2 and B2 O3 . The aluminothermic reduction of Fe2 O3 , TiO2 and B2 O3 , being highly exothermic in nature, essentially leads to a self-propagating high-temperature synthesis of the Fe-TiB2 composite. This is verified both thermodynamically and experimentally. The formation of Fe2 B, along with TiB2 , could not be avoided during the synthesis of the composite. However, it can possibly be avoided by increasing the amount of Ti or by decreasing the amount of B in the melt. The abrasive wear resistance of the composite is better than that of a high-chromium iron. The composite also possesses good high temperature stability. TiB2 particles do not undergo dissolution or coarsening during annealing at 900 ◦ C. References
Fig. 9. SEM micrograph of the worn surface of as-cast composite at a load of 19.6 N.
[1] E. Pagounis, M. Talvitie, V.K. Lindross, Influence of matrix structure on the abrasion wear resistance and toughness of a hot isostatic pressed white iron matrix composite, Metall. Mater. Trans. A 27 (1997) 4183–4191. [2] C.C. Degnan, P.H. Shipway, The incorporation of self-propagating, hightemperature synthesis-formed Fe-TiB2 into ferrous melt, Metall. Mater. Trans. A 33 (2002) 2973–2983. [3] E. Pagounis, M. Talvitie, V.K. Lindross, Influence of reinforcement volume fraction and size on the microstructure and abrasion wear resistance
76
[4]
[5]
[6]
[7] [8]
[9]
[10]
A. Anal et al. / Journal of Materials Processing Technology 172 (2006) 70–76 of hot isostatic pressed white iron matrix composite, Metall. Mater. Trans. A 27 (1997) 4171–4181. T.Z. Kattamis, T. Sugamuma, Solidification processing and tribological behavior of particulate TiC-ferrous matrix composites, Mater. Sci. Eng. A 128 (1990) 241–252. S. Skolianos, T.Z. Kattamis, M. Chen, B.V. Chambers, Cast microstructure and tribological properties of particulate TiC-reinforced Ni-base or stainless steel matrix composites, Mater. Sci. Eng. A 183 (1994) 195–204. C.C. Degnan, P.H. Shipway, J.V. Wood, Elevated temperature sliding wear behaviour of TiC-reinforced steel matrix composites, Wear 251 (2001) 1444–1451. B.S. Terry, O.S. Chinyamakobvu, Dispersion and reaction of TiB2 in liquid iron alloy, Mater. Sci. Technol. 8 (1992) 491–499. B.S. Terry, O.S. Chinyamakobvu, Carbothermic reduction of ilmenite and rutile as means of production of iron based Ti(O,C) metal matrix composites, Mater. Sci. Technol. 7 (1991) 842–848. P.P. Bandyopadhyay, Processing and characterization of plasma sprayed ceramic coatings on steel substrate, PhD thesis, Indian Institute of Technology-Kharagpur, India, 2001. O. Kubaschewski, C.B. Alcock, P.J. Spencer, Materials Thermochemistry, Pergamon press, Oxford, 1993, p. 258.
[11] D.R. Gaskell, Introduction to Metallurgical Thermodynamics, second ed., McGraw-Hill Book Company, New York, USA, 1981, pp. 585– 587. [12] T.B. Massalski, Binary Alloy Phase Diagrams, 1, Second ed., ASM International, Materials Park, OH 44073, 2001, p. 482. [13] T.B. Massalski, Binary Alloy Phase Diagrams, 2, Second ed., ASM International, Materials Park, OH 44073, 2001, p. 1785. [14] H.I. Aaronson, G. Spanos, R.A. Masamura, R.G. Vardiman, D.W. Moon, E.S.K. Menon, M.J. Hall, Sympathetic nucleation: an overview, Mater. Sci. Eng. B 32 (1995) 107–123. [15] H.E. Boyer, T.L. Gall (Eds.), Metals Handbook, desk ed., American Society for Metals, Metals Park, OH, 1985, p. 420. [16] H. Holleck, H. Leiste, W. Schneider, Significance of phase boundaries in wear resistant TiC/TiB2 materials, R an HM (1987) 149–154. [17] V.K. Rai, R. Srivastava, S.K. Nath, S. Ray, Wear in cast titanium carbide reinforced ferrous composites under dry sliding wear, Wear 231 (1991) 265–271. [18] E. Pagounis, M. Talvitie, V.K. Lindroos, Microstructure and mechanical properties of hot work tool steel matrix composites produced by hot isostatic pressing, Powder Metall. 40 (1997) 55–61. [19] Y. Geller, Tool Steels, Mir Publishers, Moscow, 1978, p. 186. [20] M.M. Khruschov, Principles of abrasive wear, Wear 69 (1974) 69–88.