- Email: [email protected]
Wear resistance of a functionally-graded aluminum matrix composite
Scripta Materialia 55 (2006) 95–98 www.actamat-journals.com
Wear resistance of a functionally-graded aluminum matrix composite Z. Humberto Melgarejo,a O. Marcelo Sua´rezb,* and Kumar Sridharanc a
Department of Mechanical Engineering, University of Puerto Rico-Mayagu¨ez, P.O. Box 9045, Mayagu¨ez 00681-9045, Puerto Rico b Department of General Engineering, University of Puerto Rico-Mayagu¨ez, P.O. Box 9044, Mayagu¨ez 00681-9044, Puerto Rico c Department of Engineering Physics, University of Wisconsin-Madison, 1500 Engineering Dr., Madison, WI 53706, USA Received 17 October 2005; revised 10 February 2006; accepted 16 March 2006 Available online 14 April 2006
Aluminum-based functionally graded composites reinforced with aluminum diboride particles were manufactured by centrifugal casting. The centrifugally cast materials showed the segregation of diboride particles and higher superficial hardness and microhardness towards the outer region of the casting. Ball-on-disk wear tests showed the smallest wear volume on the casting external zones. Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Atomic force microscopy (AFM); Metal matrix composites (MMC); Centrifugal casting; Functionally graded materials (FGM); Wear
Functionally graded advanced composite materials contain reinforcement particles whose volume fraction varies continuously from the inner to the outer sections of the part thereby providing for a controlled nonuniform microstructure with continuously changing properties [1]. This type of functionally graded material (FGM) is useful in applications where a combination of high surface wear resistance and high toughness of the interior bulk material is required. This balance of properties may not be achievable in many cases in monolithic or homogeneous materials [2]. Surface modification and coating methods can have this effect and are widely used to improve tribological properties of metallic materials, but the added surface treatment can increase the manufacturing costs. Metal matrix composites (MMCs) reinforced with ceramic or metallic particles are widely used owing to their higher specific modulus, strength and wear resistance. Furthermore MMCs have been considered as alternatives to monolithic metallic materials or conventional alloys in a number of specialized application areas [3]. Aluminum matrix composites (AMCs) have been reported to possess higher wear resistance and lower friction with an increasing volume fraction of reinforcement particles, compared to aluminum alloys without reinforcement [4]. AMCs also combine the low density of the matrix with the high hardness of the reinforcements. * Corresponding author. Tel.: +1 787 832 4040x2350; fax: +1 787 265 3816; e-mail: [email protected]
In Al–B alloys the practical insolubility of boron in solid and liquid Al confers significance to this system since these alloys, in practical terms, are Al matrix composites (AMCs) reinforced with AlB2 particles [5]. Additionally, this diboride has a higher density, 3.19 g/cm3 [6], than liquid Al (2.4 g/cm3) [7] at the semisolid composite casting temperatures (>700 °C). Centrifugal casting is an effective method to process FGMs. This type of pressure casting method enables the pouring of molten metal into the mold when the force of gravity acts on the mold assembly due to the rotational or spinning motion. The principal advantage of centrifugal casting is good mold filling combined with good microstructural control, which usually results in improved mechanical properties. However, a complete description of the particle segregation and distribution phenomena has not yet been fully given. Nevertheless, it is possible to create a composition gradient due to the difference in material density [8]. Considering the aforementioned aspects of AlB2-containing AMCs, this paper reports the wear response of FGM–AMCs reinforced with AlB2 particles produced by centrifugal casting of an initially uniform dispersion of AlB2 particles in molten aluminum. The redistribution of volume fraction of reinforcement particles was controlled by inertial forces toward the parts’ outer region during centrifugal casting. This redistribution was aided by the higher density of the AlB2 particles compared to the molten aluminum matrix. The forced segregation of hard boride particles towards the outer regions
1359-6462/$ - see front matter Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2006.03.031
96
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
of the casting by centrifugal forces provided a unique approach to improving surface hardness [9] and wear resistance of the composite, while retaining high levels of toughness in the interior regions of the part. To produce this FGM-AMC, centrifugal casting was applied to a commercial Al–5wt.%B alloy and an experimental Al–4wt.%B and 2 wt.% Mg alloy. The centrifugal casting unit used in this study was spring-driven and used cylindrical flask sizes up to 88.9 mm diameter and 127 mm in length. Additionally, a preheated transfer scoop was utilized to pour molten material, which, subject to centrifugal forces, is driven into a mold prepared by investment casting. The centrifugal casting process was carried out with an initial (maximum) rotational velocity of 200 rpm while the pouring temperature of the material was set at 750 °C. The cylindrical samples obtained were 16 mm in diameter and 20 mm in length (Fig. 1). This equipment is ideally suited to small-scale laboratory experiments (50–60 g of charge material) of developmental alloys. Microstructure, volume fraction of reinforcement particles, superficial Rockwell hardness (HR15W), Vickers microhardness (HV50) and wear rates were evaluated for the internal and external zones of the final centrifugally cast piece, as indicated in Figure 1. The microstructural details at these locations were observed using a Nikon Epiphot 2 optical microscope. Quantitative image analysis was used to evaluate the distribution of the reinforcement particles for Al–B and Al–Mg–B alloys in three conditions: as-received condition, gravity cast condition and internal and external zones of the centrifugally cast samples. These results were correlated with the values of superficial Rockwell hardness and Vickers microhardness measured for those composites. Unlubricated ball-on-disk configuration wear tests were conducted on the external and internal zones of the centrifugally cast AMCs and on stationary (gravity) cast specimens (disks). The wear tests were conducted at a constant sliding speed, 0.0066 m s1. The counter-surface (pin) used was a 3.175 mm diameter ball of AISI 410 martensitic stainless steel. The duration of the wear test and the applied normal load were maintained at 45 min and 0.15 N (15 g), respectively, for all tests. The total sliding distance under these conditions was 18.5 m. The wear volume was calculated from topographical characterization of the wear tracks using a Nikon SMZ 1500 stereoscope, the optical microscope, and a Nanosurf easyScan atomic force microscope (AFM).
Figure 1. Mapped longitudinal zone of centrifugally cast specimen.
Figure 2 shows the resulting microstructure with AlB2 reinforcements embedded in the aluminum matrix for a binary Al–5wt.%B composite. Figure 3 presents the distribution of the reinforcement (AlB2) volume fraction obtained by quantitative analysis. Naturally, the as-received and gravity cast AMCs exhibited a uniform distribution of the reinforcement particles. The composite samples produced by centrifugal casting showed higher densities of reinforcement particles in the outer regions compared to the inner regions of the casting. This occurs since the dispersed particles are segregated by centrifugal forces and the thickness of the particle-rich region is strongly influenced by the speed of rotation, the local solidification time and the density difference between the base alloy and the reinforcement [10]. In addition, the Al–Mg–B composite exhibited more significant reinforcement particles segregation than Al–B composite. Magnesium decreases the molten aluminum alloy density and consequently provides fluidity to the melt. On the other hand, it has already been established that higher boron levels, i.e. higher diboride volume fractions, increases the viscosity of the semisolid material [9]. This higher viscosity hampers the movement the diboride particles, which causes a lesser gradient of diboride volume fraction from the inner to the outer sections of the cast. Figure 4 shows the variation of the superficial Rockwell hardness on cross-sections of these AMCs measured for the aforementioned experimental conditions. The results revealed that the value of superficial
Figure 2. Micrograph of the microstructure of a centrifugally cast Al– 5wt.%B composite.
Figure 3. Measured volume fraction of reinforcement particles in the AMC samples investigated in this study.
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
Figure 4. Rockwell superficial hardness (HR15W) measured in the AMC samples investigated in this study.
hardness increases as a function of distance from the internal zone to the external zone in the experimental Al–4wt.%B–2wt.%Mg composite. The Al–5wt.%B alloy did not demonstrate this behavior because the reinforcement particle gradient is not as apparent as the one observed in the Al–Mg–B composite. Moreover, additional electron microprobe analysis indicated that, when Mg is present in the composite, it diffuses in the AlB2 particles to substitute Al atoms and forms an AlxMg1xB2 compound. Additionally, it is believed that this AlxMg1xB2 compound is harder than AlB2. This assumption is being corroborated at this moment by means of nanoindentation tests. Results of Vickers microhardness analyses selectively carried out on the Al matrix and reported in Figure 5 support the superficial Rockwell hardness measurements. One could assume that the hardness of the matrix is not affected by the reinforcement particles. However, this behavior is not often observed since the values of Young’s modulus and the thermal expansion coefficients in ceramic and intermetallic compound particles are different from those of the metal matrix. Hence, the residual strain energy caused by the stress field present in the metal matrix of the FGMs, depends on the composition gradient. The mutual relation among hardness, volume fraction of particles and strain energy can satisfy the supposition that the hardness and strain energy are proportional to the composition gradient [8]. The addition of magnesium to the aluminum matrix has significant effects on their mechanical and microstructural properties compared with Al–B composites. This occurs because magnesium has a very high solid solubility in aluminum and the rate of precipitation of the supersaturated solid solution is low. This explains
Figure 5. Vickers microhardness (HV50) values measured for the matrices of the different AMCs investigated in this study.
97
the higher values of microhardness in the matrix of Al–Mg–B composites with respect to the matrix hardness of Al–B composites. These Vickers microhardness values are in agreement with prior research. The microhardness result of Al–B casting can be attributed to its larger reinforcement particle size regarding Al–B centrifugally cast composites. This particle size can be diminished owing to interaction between particles during centrifugal casting. Even though the effect of reinforcement particle size was not dealt with in this study, its influence has already been analyzed in the literature. The topographical aspects, as well as the dimensions of the pin-on-disk tracks such as perimeter, width and depth, are presented in Figure 6. These parameters allowed for the calculation of the wear volume on each sample. Figure 7 shows the volumetric wear results obtained from gravity-cast samples as well as on the internal and external zones of the centrifugally cast samples of the Al–5wt.%B and Al–4wt.%B–2wt.%Mg. The smallest wear volume measured was observed in the external zone of the FGMs. This is in agreement with the observed higher volume fraction of boride phases in these outer regions. On the other hand, the internal zone of the FGMs (more depleted of AlB2 particles) resulted in higher wear volume. The gravity-cast sample displayed a wear volume loss in between those exhibited by the outer and inner regions of the sample. Furthermore, the Al– 4wt.%B–2wt.%Mg composite exhibited a lower wear volume in the external zone than the centrifugally cast Al–5wt.%B alloy. This effect is attributed to solid solution strengthening and precipitation hardening of metastable Mg-containing phases. Table 1 shows the wear coefficients according to Eq. (1), calculated for the samples exposed to unlubricated pin-on-disk wear tests. W ð1Þ k¼ FN s where W is the wear volume (mm3); FN, the applied load (N) and S is the sliding distance (m). Wear test results, and wear track analysis were consistent with the microstructural gradient observed in these alloys. In effect, the higher volume density of reinforcing boride particles in the outer regions (external zones) of the centrifugally cast samples translated into a higher overall wear resistance on those regions. On the other hand, the internal regions (internal zones) were fairly depleted of boride reinforcement particles and, thus, were subject to high wear rates. Finally, as expected, the control sample (gravity-cast composite) displayed an intermediate behavior as the volume fraction of diborides was evenly distributed throughout the composite mass. The beneficial effect of Mg in the wear resistance of these AMCs requires further experimentation and is the focus of ongoing research. Initial experimentation using differential calorimetric techniques proved the precipitation of metastable phases in the Mg-containing composites. The precipitates are advantageous in enhancing wear resistance of the composites (precipitation hardening of the matrix). This result has important practical implications because, by controlling the ageing process of these composites, their hardness and wear
98
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
Figure 6. Images of wear tracks resulting from pin-on-disk wear tests: (a) wear track of Al–4wt.%B–2wt.%Mg external zone of FGM; (b) track width of Al–5wt.%B casting; (c) track depth of Al–5wt.%B internal zone of FGM.
and, thus, were subject to higher wear rates. As expected, the control sample (gravity-cast composite) displayed an intermediate behavior as the volume fraction of diborides is evenly distributed throughout the composite mass. On the other hand Al–B composites did not fully display this behavior. Superficial hardness and microhardness were lower than Al–Mg–B composite and the reinforcement particles gradient was not particularly relevant.
Figure 7. Wear volume for the various AMC samples after undergoing pin-on-disk wear tests under identical conditions. Table 1. Calculated wear coefficient for the various AMC samples Wear coefficient (mm3/N m)
Casting Internal zone—200 rpm External zone—200 rpm
Al–5wt.%B
Al–4wt.%B–2wt.%Mg
0.00374 0.00590 0.00249
0.00472 0.00922 0.00179
resistance can be conveniently tailored, as previously reported in a research on AMCs containing copper [5]. In general, reinforcing AlB2 particles (with higher bulk density than liquid aluminum) segregate towards the surface regions as a result of the inertial forces created during the centrifugal casting process. Additionally, hardness values, wear test results, and wear track analysis were consistent with the microstructural gradient observed in Al–B–Mg composites. In effect, the higher volume density of reinforcing boride particles in the outer regions of the centrifugally cast samples translates into a higher hardness and higher overall wear resistance on those regions. On the other hand, the internal regions were fairly depleted of boride reinforcement particles
The authors would like to acknowledge the generous financial contribution of Puerto Rico EPSCoR. This material is partially based upon work supported by the National Science Foundation under Grant No. 0351449. The Al–B master alloys were kindly donated by Milward Alloys, Inc., Lockport, New York. Grant No. N000140310540 awarded by the US Navy Office of Naval Research allowed for the AFM instrumentation. [1] G.H. Paulino, J.H. Kim, J. Eng. Fract. Mech. 71 (2004) 1907. [2] S. Nai, M. Gupta, C. Lim, Mater. Sci. Technol. 20 (2004) 57–67. [3] J.R. Gomes, A.S. Miranda, D. Soares, A.E. Dias, L.A. Rocha, Ceram. Trans. 114 (2000) 579. [4] M.H. Korkut, Mater. Sci. Technol. 20 (2004) 73–81. [5] O.M. Sua´rez, J. Mech. Behav. Mater. 11 (2001) 225. [6] R. David, Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL, 1998. [7] A. Rai, D. Lee, K. Park, M.R. Zachariah, J. Phys. Chem. 108 (2004) 14793–14795. [8] Y. Watanabe, Y. Fukui, Recent Res. Dev. Metall. Mater. Sci. 4 (2000) 51–93. [9] N.B. Duque, Z.H. Melgarejo, O.M. Sua´rez, J. Mater. Charact. 55 (2005) 167–171. [10] C.G. Kang, P.K. Rohatgi, Metall. Mater. Trans. 27B (1996) 277–285.
Wear resistance of a functionally-graded aluminum matrix composite Z. Humberto Melgarejo,a O. Marcelo Sua´rezb,* and Kumar Sridharanc a
Department of Mechanical Engineering, University of Puerto Rico-Mayagu¨ez, P.O. Box 9045, Mayagu¨ez 00681-9045, Puerto Rico b Department of General Engineering, University of Puerto Rico-Mayagu¨ez, P.O. Box 9044, Mayagu¨ez 00681-9044, Puerto Rico c Department of Engineering Physics, University of Wisconsin-Madison, 1500 Engineering Dr., Madison, WI 53706, USA Received 17 October 2005; revised 10 February 2006; accepted 16 March 2006 Available online 14 April 2006
Aluminum-based functionally graded composites reinforced with aluminum diboride particles were manufactured by centrifugal casting. The centrifugally cast materials showed the segregation of diboride particles and higher superficial hardness and microhardness towards the outer region of the casting. Ball-on-disk wear tests showed the smallest wear volume on the casting external zones. Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Atomic force microscopy (AFM); Metal matrix composites (MMC); Centrifugal casting; Functionally graded materials (FGM); Wear
Functionally graded advanced composite materials contain reinforcement particles whose volume fraction varies continuously from the inner to the outer sections of the part thereby providing for a controlled nonuniform microstructure with continuously changing properties [1]. This type of functionally graded material (FGM) is useful in applications where a combination of high surface wear resistance and high toughness of the interior bulk material is required. This balance of properties may not be achievable in many cases in monolithic or homogeneous materials [2]. Surface modification and coating methods can have this effect and are widely used to improve tribological properties of metallic materials, but the added surface treatment can increase the manufacturing costs. Metal matrix composites (MMCs) reinforced with ceramic or metallic particles are widely used owing to their higher specific modulus, strength and wear resistance. Furthermore MMCs have been considered as alternatives to monolithic metallic materials or conventional alloys in a number of specialized application areas [3]. Aluminum matrix composites (AMCs) have been reported to possess higher wear resistance and lower friction with an increasing volume fraction of reinforcement particles, compared to aluminum alloys without reinforcement [4]. AMCs also combine the low density of the matrix with the high hardness of the reinforcements. * Corresponding author. Tel.: +1 787 832 4040x2350; fax: +1 787 265 3816; e-mail: [email protected]
In Al–B alloys the practical insolubility of boron in solid and liquid Al confers significance to this system since these alloys, in practical terms, are Al matrix composites (AMCs) reinforced with AlB2 particles [5]. Additionally, this diboride has a higher density, 3.19 g/cm3 [6], than liquid Al (2.4 g/cm3) [7] at the semisolid composite casting temperatures (>700 °C). Centrifugal casting is an effective method to process FGMs. This type of pressure casting method enables the pouring of molten metal into the mold when the force of gravity acts on the mold assembly due to the rotational or spinning motion. The principal advantage of centrifugal casting is good mold filling combined with good microstructural control, which usually results in improved mechanical properties. However, a complete description of the particle segregation and distribution phenomena has not yet been fully given. Nevertheless, it is possible to create a composition gradient due to the difference in material density [8]. Considering the aforementioned aspects of AlB2-containing AMCs, this paper reports the wear response of FGM–AMCs reinforced with AlB2 particles produced by centrifugal casting of an initially uniform dispersion of AlB2 particles in molten aluminum. The redistribution of volume fraction of reinforcement particles was controlled by inertial forces toward the parts’ outer region during centrifugal casting. This redistribution was aided by the higher density of the AlB2 particles compared to the molten aluminum matrix. The forced segregation of hard boride particles towards the outer regions
1359-6462/$ - see front matter Ó 2006 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2006.03.031
96
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
of the casting by centrifugal forces provided a unique approach to improving surface hardness [9] and wear resistance of the composite, while retaining high levels of toughness in the interior regions of the part. To produce this FGM-AMC, centrifugal casting was applied to a commercial Al–5wt.%B alloy and an experimental Al–4wt.%B and 2 wt.% Mg alloy. The centrifugal casting unit used in this study was spring-driven and used cylindrical flask sizes up to 88.9 mm diameter and 127 mm in length. Additionally, a preheated transfer scoop was utilized to pour molten material, which, subject to centrifugal forces, is driven into a mold prepared by investment casting. The centrifugal casting process was carried out with an initial (maximum) rotational velocity of 200 rpm while the pouring temperature of the material was set at 750 °C. The cylindrical samples obtained were 16 mm in diameter and 20 mm in length (Fig. 1). This equipment is ideally suited to small-scale laboratory experiments (50–60 g of charge material) of developmental alloys. Microstructure, volume fraction of reinforcement particles, superficial Rockwell hardness (HR15W), Vickers microhardness (HV50) and wear rates were evaluated for the internal and external zones of the final centrifugally cast piece, as indicated in Figure 1. The microstructural details at these locations were observed using a Nikon Epiphot 2 optical microscope. Quantitative image analysis was used to evaluate the distribution of the reinforcement particles for Al–B and Al–Mg–B alloys in three conditions: as-received condition, gravity cast condition and internal and external zones of the centrifugally cast samples. These results were correlated with the values of superficial Rockwell hardness and Vickers microhardness measured for those composites. Unlubricated ball-on-disk configuration wear tests were conducted on the external and internal zones of the centrifugally cast AMCs and on stationary (gravity) cast specimens (disks). The wear tests were conducted at a constant sliding speed, 0.0066 m s1. The counter-surface (pin) used was a 3.175 mm diameter ball of AISI 410 martensitic stainless steel. The duration of the wear test and the applied normal load were maintained at 45 min and 0.15 N (15 g), respectively, for all tests. The total sliding distance under these conditions was 18.5 m. The wear volume was calculated from topographical characterization of the wear tracks using a Nikon SMZ 1500 stereoscope, the optical microscope, and a Nanosurf easyScan atomic force microscope (AFM).
Figure 1. Mapped longitudinal zone of centrifugally cast specimen.
Figure 2 shows the resulting microstructure with AlB2 reinforcements embedded in the aluminum matrix for a binary Al–5wt.%B composite. Figure 3 presents the distribution of the reinforcement (AlB2) volume fraction obtained by quantitative analysis. Naturally, the as-received and gravity cast AMCs exhibited a uniform distribution of the reinforcement particles. The composite samples produced by centrifugal casting showed higher densities of reinforcement particles in the outer regions compared to the inner regions of the casting. This occurs since the dispersed particles are segregated by centrifugal forces and the thickness of the particle-rich region is strongly influenced by the speed of rotation, the local solidification time and the density difference between the base alloy and the reinforcement [10]. In addition, the Al–Mg–B composite exhibited more significant reinforcement particles segregation than Al–B composite. Magnesium decreases the molten aluminum alloy density and consequently provides fluidity to the melt. On the other hand, it has already been established that higher boron levels, i.e. higher diboride volume fractions, increases the viscosity of the semisolid material [9]. This higher viscosity hampers the movement the diboride particles, which causes a lesser gradient of diboride volume fraction from the inner to the outer sections of the cast. Figure 4 shows the variation of the superficial Rockwell hardness on cross-sections of these AMCs measured for the aforementioned experimental conditions. The results revealed that the value of superficial
Figure 2. Micrograph of the microstructure of a centrifugally cast Al– 5wt.%B composite.
Figure 3. Measured volume fraction of reinforcement particles in the AMC samples investigated in this study.
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
Figure 4. Rockwell superficial hardness (HR15W) measured in the AMC samples investigated in this study.
hardness increases as a function of distance from the internal zone to the external zone in the experimental Al–4wt.%B–2wt.%Mg composite. The Al–5wt.%B alloy did not demonstrate this behavior because the reinforcement particle gradient is not as apparent as the one observed in the Al–Mg–B composite. Moreover, additional electron microprobe analysis indicated that, when Mg is present in the composite, it diffuses in the AlB2 particles to substitute Al atoms and forms an AlxMg1xB2 compound. Additionally, it is believed that this AlxMg1xB2 compound is harder than AlB2. This assumption is being corroborated at this moment by means of nanoindentation tests. Results of Vickers microhardness analyses selectively carried out on the Al matrix and reported in Figure 5 support the superficial Rockwell hardness measurements. One could assume that the hardness of the matrix is not affected by the reinforcement particles. However, this behavior is not often observed since the values of Young’s modulus and the thermal expansion coefficients in ceramic and intermetallic compound particles are different from those of the metal matrix. Hence, the residual strain energy caused by the stress field present in the metal matrix of the FGMs, depends on the composition gradient. The mutual relation among hardness, volume fraction of particles and strain energy can satisfy the supposition that the hardness and strain energy are proportional to the composition gradient [8]. The addition of magnesium to the aluminum matrix has significant effects on their mechanical and microstructural properties compared with Al–B composites. This occurs because magnesium has a very high solid solubility in aluminum and the rate of precipitation of the supersaturated solid solution is low. This explains
Figure 5. Vickers microhardness (HV50) values measured for the matrices of the different AMCs investigated in this study.
97
the higher values of microhardness in the matrix of Al–Mg–B composites with respect to the matrix hardness of Al–B composites. These Vickers microhardness values are in agreement with prior research. The microhardness result of Al–B casting can be attributed to its larger reinforcement particle size regarding Al–B centrifugally cast composites. This particle size can be diminished owing to interaction between particles during centrifugal casting. Even though the effect of reinforcement particle size was not dealt with in this study, its influence has already been analyzed in the literature. The topographical aspects, as well as the dimensions of the pin-on-disk tracks such as perimeter, width and depth, are presented in Figure 6. These parameters allowed for the calculation of the wear volume on each sample. Figure 7 shows the volumetric wear results obtained from gravity-cast samples as well as on the internal and external zones of the centrifugally cast samples of the Al–5wt.%B and Al–4wt.%B–2wt.%Mg. The smallest wear volume measured was observed in the external zone of the FGMs. This is in agreement with the observed higher volume fraction of boride phases in these outer regions. On the other hand, the internal zone of the FGMs (more depleted of AlB2 particles) resulted in higher wear volume. The gravity-cast sample displayed a wear volume loss in between those exhibited by the outer and inner regions of the sample. Furthermore, the Al– 4wt.%B–2wt.%Mg composite exhibited a lower wear volume in the external zone than the centrifugally cast Al–5wt.%B alloy. This effect is attributed to solid solution strengthening and precipitation hardening of metastable Mg-containing phases. Table 1 shows the wear coefficients according to Eq. (1), calculated for the samples exposed to unlubricated pin-on-disk wear tests. W ð1Þ k¼ FN s where W is the wear volume (mm3); FN, the applied load (N) and S is the sliding distance (m). Wear test results, and wear track analysis were consistent with the microstructural gradient observed in these alloys. In effect, the higher volume density of reinforcing boride particles in the outer regions (external zones) of the centrifugally cast samples translated into a higher overall wear resistance on those regions. On the other hand, the internal regions (internal zones) were fairly depleted of boride reinforcement particles and, thus, were subject to high wear rates. Finally, as expected, the control sample (gravity-cast composite) displayed an intermediate behavior as the volume fraction of diborides was evenly distributed throughout the composite mass. The beneficial effect of Mg in the wear resistance of these AMCs requires further experimentation and is the focus of ongoing research. Initial experimentation using differential calorimetric techniques proved the precipitation of metastable phases in the Mg-containing composites. The precipitates are advantageous in enhancing wear resistance of the composites (precipitation hardening of the matrix). This result has important practical implications because, by controlling the ageing process of these composites, their hardness and wear
98
Z. H. Melgarejo et al. / Scripta Materialia 55 (2006) 95–98
Figure 6. Images of wear tracks resulting from pin-on-disk wear tests: (a) wear track of Al–4wt.%B–2wt.%Mg external zone of FGM; (b) track width of Al–5wt.%B casting; (c) track depth of Al–5wt.%B internal zone of FGM.
and, thus, were subject to higher wear rates. As expected, the control sample (gravity-cast composite) displayed an intermediate behavior as the volume fraction of diborides is evenly distributed throughout the composite mass. On the other hand Al–B composites did not fully display this behavior. Superficial hardness and microhardness were lower than Al–Mg–B composite and the reinforcement particles gradient was not particularly relevant.
Figure 7. Wear volume for the various AMC samples after undergoing pin-on-disk wear tests under identical conditions. Table 1. Calculated wear coefficient for the various AMC samples Wear coefficient (mm3/N m)
Casting Internal zone—200 rpm External zone—200 rpm
Al–5wt.%B
Al–4wt.%B–2wt.%Mg
0.00374 0.00590 0.00249
0.00472 0.00922 0.00179
resistance can be conveniently tailored, as previously reported in a research on AMCs containing copper [5]. In general, reinforcing AlB2 particles (with higher bulk density than liquid aluminum) segregate towards the surface regions as a result of the inertial forces created during the centrifugal casting process. Additionally, hardness values, wear test results, and wear track analysis were consistent with the microstructural gradient observed in Al–B–Mg composites. In effect, the higher volume density of reinforcing boride particles in the outer regions of the centrifugally cast samples translates into a higher hardness and higher overall wear resistance on those regions. On the other hand, the internal regions were fairly depleted of boride reinforcement particles
The authors would like to acknowledge the generous financial contribution of Puerto Rico EPSCoR. This material is partially based upon work supported by the National Science Foundation under Grant No. 0351449. The Al–B master alloys were kindly donated by Milward Alloys, Inc., Lockport, New York. Grant No. N000140310540 awarded by the US Navy Office of Naval Research allowed for the AFM instrumentation. [1] G.H. Paulino, J.H. Kim, J. Eng. Fract. Mech. 71 (2004) 1907. [2] S. Nai, M. Gupta, C. Lim, Mater. Sci. Technol. 20 (2004) 57–67. [3] J.R. Gomes, A.S. Miranda, D. Soares, A.E. Dias, L.A. Rocha, Ceram. Trans. 114 (2000) 579. [4] M.H. Korkut, Mater. Sci. Technol. 20 (2004) 73–81. [5] O.M. Sua´rez, J. Mech. Behav. Mater. 11 (2001) 225. [6] R. David, Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL, 1998. [7] A. Rai, D. Lee, K. Park, M.R. Zachariah, J. Phys. Chem. 108 (2004) 14793–14795. [8] Y. Watanabe, Y. Fukui, Recent Res. Dev. Metall. Mater. Sci. 4 (2000) 51–93. [9] N.B. Duque, Z.H. Melgarejo, O.M. Sua´rez, J. Mater. Charact. 55 (2005) 167–171. [10] C.G. Kang, P.K. Rohatgi, Metall. Mater. Trans. 27B (1996) 277–285.