Bonfring International J ournal of Industrial Engineering and Management Science, Vol. 2, No.
4, December 2012 131
Investigation of Effect of Thermal Stresses on Damping and Thermal Expansion Behavior of Al/Al 2 O 3 Metal Matrix Composites B.V. Madhu, K. Pralhada Rao and D.P. Girish Abstract--- The present investigation is based on the effect of thermal residual stresses on the both thermal expansion and damping behavior of Al/Al 2 O 3 Metal Matrix Composites (MMCs) processed by compo casting. Thermal expansion and damping properties have been studied experimentally as a function of temperature over a temperature range from room temperature to 400C both in the heating and cooling cycles. Result shows the thermal expansion studies exhibited some residual strain, which increased with the increase in the weight % of the Al 2 O 3 . The damping capacity of both the composites and matrix alloy is found to increase with the increase in temperature during the heating cycle, whereas in the cooling cycle damping behavior exhibited maxima which became more pronounced with the increase in the weight percentage of the Al 2 O 3 . The damping maximum height decreases if the matrix strength is increased by increasing Al 2 O 3 . The appearance of the maxima may be linked with dislocation generation and motion as a result of plastic deformation of the matrix at the Al/Al 2 O 3 interface. This phenomenon is attributed to the thermal stresses generated as a result of coefficient of thermal expansion mismatch between the composite constituent phases. Keywords--- Compocasting, CTE, Damping, PLC, TMA
I. INTRODUCTION ETAL Matrix Composites (MMCs) offers to modify the damping behavior, as well as other physical and mechanical properties of metals and alloys by combining them with nonmetallic phases. MMCs processing provides the possibility of tailoring the resultant damping properties of MMCs by selecting high damping reinforcing materials and varying reinforcement volume fraction and geometries. In other hand Al based MMCs offer designers many added benefits as they are particularly suited for applications requiring high specific strength at elevated temperatures, good
B.V. Madhu, Research Scholar, Mechanical Engineering Department, Jawaharlal Nehru Technological University, Anantapur(AP), India, E- mail:bvmadhu2000@gmail.com K. Pralhada Rao, Vice - Principal & Professor, Mechanical Engineering Department, JNTU College of Engineering, Anantapur(AP), India, E-mail: drkprao1@yahoo.com D.P. Girish, Professor, Mechanical Engineering Department, Government Engineering College, Ramanagara, Karnataka, India, E-mail: omganesha16@yahoo.co.in
DOI: 10.9756/BIJIEMS.1809
structural rigidity, dimensional stability [1-5] and low Coefficient of Thermal Expansion (CTE) [6]. Low CTE and high damping capacity are desirable for applications such as electronic heat sinks and space structures. The MMCs in structural applications are dependent on their proper and complete characterization under various conditions of mechanical and thermal loading. One of the important aspects of MMCs is the influence of thermal residual stresses on the properties of the materials. When MMCs are fabricated at a certain high temperature and cooled to the room temperature, residual thermal stresses are induced into the matrix and reinforcement because of the significant difference between the CTE of the two constituents. The residual stresses would introduce some undesirable effects on the physical and thermal properties of the composites. Several publications on experimental works and numerical computations of thermal stresses have been found in the literature [7, 8]. However, due to the complexity of the MMCs, a better understanding of the residual stresses in these materials is essential. Earlier research works show that Al 2 O 3 particle reinforced Al MMCs offer lightweight, low CTE [6], excellent mechanical properties [9,10], good structural stability at high temperature [11,12], good wear [13], and corrosion resistance [14]. The present investigation was undertaken with the objectives of systematically studying the damping behavior, CTE and thermal stresses developed due to thermal and damping. II. EXPERIMENTAL PROCEDURE 2.1 Material Preparation The matrix material used in this study, Al6061 and Al 2 O 3
particulates of 30-50 m were used as the reinforcement. The chemical compositions of both matrix and reinforcement were given in Table I. Liquid metallurgy technique was used to fabricate the composite materials in which the Al 2 O 3 particles were introduced into the molten metal pool through a vortex created in the melt by the use of an alumina-coated stainless steel stirrer. Al 2 O 3 content in the composites was varied from 5 to 15% in steps of 5% by weight. Table I shows the chemical composition of Al6061 and Al 2 O 3 .
2.2 5m usi and fin ace me 50 Me US pla top det (LV sam env adj dim tem ver sof the 2.3 x pre An dam piv pro sub am sim cla coo fle sam dis cor len mm to 10 Table I: C 2 CTE Measur The specimen mm x 5mm dim ing series of S d then polishe nish. The speci etone separate easurements w 0C at the rat echanical Anal SA) both in h aced on a quar p of the sample tected using a VDT) attached mple stage a vironment du jacent to the s mensional cha mperature. The rsus temperat ftware was use e composites. 3 Damping Me Specimens we 2mm dimensi esented under nalyzer (DMA mping capacity vot, LVDT, e ogrammer inte bjected to a mplitude and multaneously. amps and enclo oling capabil xural strain to mple was me splaced by 25 rresponding to ngth measured m. The temper 500C both in C/min. Chemical Comp Constituents Mg Si Fe Cu Ti Cr Zn Mn Be V Al rement ns for CTE tes mensions. The iC papers of 1 ed with 3m d imens were wa ely and dried. were made in te of 5C/min lyzer Equipme heating and co rtz stage and a e. The thermal a Linear Vari d to the probe and the prob uring measu sample picks u anges can be e data were o ture curves. S ed to compute easurement ere machined a ions following CTE measure A model 983, y includes sam electro-magnet erfaced with a flexural sin the resultant-b The sample w osed in a cham lity. The ele o the specimen easured by L 50 m peak o a maximum d between the rature of the sp n the heating a position of Al6 Composition i % 0.92 0.76 0.28 0.22 0.10 0.07 0.06 0.04 0.003 0.01 Balance ting were mac specimen surfa 100, 200, 400, diamond paste ashed using di Percent Linea the temperat using a Com ent (model TM ooling cycles. a moveable pr expansion of t iable Differen e. The furnace e provides h urements. The up the temper e obtained as obtained in th Standard TMA the CTE of th and polished to g according t ement. A Dyn Dupont) use mple arms and tic driver and a computer. Th nusoidal strain bending stress was held betw mber that prov ectro-magnetic n and the resul LVDT. The to peak at t strain of 2.6 x clamps was a pecimen was v nd cooling cyc and Management ISSN 2277 Science, Vol. 2, N - 6061 Alloy in chined to 10mm faces were grou 600, & 1000 g e to obtain a f istilled water a ar Change (PL ture range 25 mmercial Therm MA 2940, Dupo The sample w robe was kept the specimen w ntial Transform e surrounding high temperat e thermocou rature so that s a function he form of P A data analy he base alloy a o 70mm x 10m to the proced amic Mechani d for measur d clamps, flex d a temperat he specimen w n with const s was measu ween the two e vides heating a driver appl tant stress on specimens w the drive cla x10 -4 . The sam approximately varied from 25 cles at the rate -5056 | 2012 Bo m x und grit fine and LC) to mal ont, was on was mer the ture uple the of LC ysis and mm dure ical ring ure ture was tant ured end and lies the were amp mple 38 5C e of 3.1 The The tempera 6061 M reasona at 500 curve fo in temp parabol heating strain, w Al 2 O 3 unreinfo The are increase
Figu Temp Coo The as well While C % of tempera 3.2 Da Dam base all are sho from th of the A from 2 margina found t from 30 in temp unreinfo maxima closer o No. 4, December 20012 132 ermal Expansio results of CT ature for differ MMCs is sho ably good agree C for the maj or the heating c perature, whi ic decrease w and cooling which increase reinforcement forced matrix a ea of the hyste e in the Al 2 O 3 r
onfring ure 1: Percent L perature Varied oling Cycle for variation of C as the unrein CTE of the com Al 2 O 3 , it m ature. amping Capacit mping capacity loy and the co own in Fig. 3. he curves that a Al 2 O 3 reinforc 25-150C, da ally with increa to increase rap 00-500C it inc perature. Also forced matrix a (peak) is obs observation of III. RESU on Properties TE expressed rent weight % own in Fig.1. ement with a m ority of the sp cycle shows lin le for the co with the decrea curves exhibit es with increa . The residua alloy is found eresis between reinforcement. Linear Change d at the Rate of r the Al 2 O 3 Rei CTE with temp nforced matrix mposite decrea moderately inc ty y (tan ) verses omposites for h Several intere are found to b ced Al compos amping capac ase in tempera pidly with inc creases mildly o, in the hea alloy and r served at 500 Fig. 3 reveals ULTS as PLC as a of reinforceme The PLC d maximum varia pecimens teste near increase w ooling cycle ase in tempera t some hystere se in the weig al strain in t to have a mini n the curves in . (PLC) as a Fu f 5 C/min in H inforced Al Co perature of the alloy is show ses with increa creases with temperature c heating and co esting trends a e the character sites. In the he city seems ature, from 150 crease in temp y but linearly w ating cycle, fo reinforced co C. In the coo some interestin function of ent in the Al data showed ation of 0.05 ed. The PLC with increase it shows a ature. These esis residual ght % of the the case of imum value. ncreases with unction of Heating and omposites e composites wn in Fig. 2. ase in weight increase in urves for the ooling cycles are observed ristic feature eating cycle, to increase 0-300 C it is perature, and with increase for both the mposites, a ling cycle, a ng details on Bonfring International J ournal of Industrial Engineering and Management Science, Vol. 2, No. 4, December 2012 133 the hysteresis behaviour and the maxima (peak) phenomena of the composites exhibited between the heating and cooling curves. In the cooling cycle, different nature of curves is observed. Initially, from 500-400C the damping capacity marginally decreases, from 400-250C it increases, from 250- 25C it steeply decreases with decrease in temperature. Overall, the damping capacity and the area of the hysteresis curve increase with increase in reinforcement. Careful observation of the damping capacity data in the 225-250 C range shows a peak. The damping capacity of the composites as well as the matrix alloy has been found to increase with the increase in temperature and increase in reinforcement.
Figure 2: Average Coefficient of Thermal Expansion as a Function of Temperature Varied at 5C/min in Heating and Cooling Cycles for the Al 2 O 3 Reinforced Al Composites
Figure 3: Damping Capacity (tan ) at 0.1 Hz of as-cast Al 6061 Alloy and Al 2 O 3 Reinforced Al Composites in Heating and Cooling Cycles at 10C/min IV. DISCUSSION 4.1 Thermal Behavior During cooling, the metal matrix contracts more than that of the ceramic particles, hence high-density dislocation is developed around the particles as a micro-plastic zone. The increase in the dislocation activity results in an increase of the thermal as well as the mechanical property loss [15], [16]. The dislocation near the interface has been observed from transmission electron microscopy [17]. The PLC verses temperature curves for the composites exhibit some hysteresis strain in the cooling cycle. The computed values of the percentage area covered by the heating and cooling curves are 5.89, 8.26, 16.7 and 22.7 in case of base alloy, 5%, 10%, and 15% Al 2 O 3 reinforced composites respectively. When the ceramic weight % is small, it may be assumed that each ceramic particulate is surrounded by the matrix phase, and the average inter-particle spacing is not influenced by the presence of the ceramic phase with lower CTE [18]. The ceramic particles are in hydrostatic state and their surrounding matrix phase suffers a compressive radial stress. On the other hand, in case of higher weight % of reinforcement, the average inter-particle spacing is significantly influenced. However, the Al matrix alloy phase with a higher CTE should undergo shrinking on cooling, resulting in a tensile residual stress development. During contorted secondary heating and cooling cycles, the matrix alloy covers the particulate and shrinks. The result of this is the hysteresis as observed between the heating and cooling curves. The thermal strain depends on the difference in the CTE between the particle (p) and the matrix (m) and the temperature difference T. The thermal strain [19] is given by T = T =(m-p) (T-T0) (1) where T and T0 are the melting and room temperatures respectively. According to Hookes law, elastic stress strain relation is given by: th =E T/(1-) (2) where E is the stiffness and is the Poissons ratio of the matrix. It is assumed that the thermal strain is purely elastic and the stress changes linearly with temperature which is given by combining Eqs (1) and (2): th =E (m - p) (T-T0)/(1-) (3) th =E (m - p) T/(1-) (4) For m =26 X 10 -6 /C, p =2.6 X 10 -6 /C, E =70 GPa and =0.3 Equation (3) gives th =2.34 MPa X T (5) For 1C change in temperature, the computed value of thermal stresses is 2.34 MPa at the interface between the matrix and the reinforcement, and this value is in good agreement with those obtained in other studies [20].
ISSN 2277-5056 | 2012 Bonfring Bonfring International J ournal of Industrial Engineering and Management Science, Vol. 2, No. 4, December 2012 134 4.2 Damping Behaviour Al/Al 2 O 3 composites have been developed with higher stiffness (E) and lower density (). (E/) 1/2 also represents the velocity of the elastic waves in the material. Thus, higher specific stiffness results in a higher natural frequency for the components [21], which is one of the reasons for increase in the damping capacity of the composites. Perez et al. [22] have reported that the dislocation concentration found in Al matrices of ceramic particulate-reinforced MMCs is in the order of 1013 to 1014 cm -2 . Kim [18] has reported that if the heat dissipation were taken into account as part of the plastic work, the predicted dislocation densities of the elastoplaticity model would have been in reasonable agreement with the measured dislocation densities of 109 to 1010 cm -2 . These dislocations are generated to accommodate the residual strain around the particulate; but the dislocation concentration near the reinforcement-matrix interface is larger and it decreases with increase in distance from the interface. The thermal stresses in MMCs have been already studied by internal friction or damping capacity measurement [23]. The existence of the damping maxima is particularly important since this maximum may be related to the thermal stresses, which are generated in the composites. In the absence of thermal stresses, the dislocation can vibrate around its equilibrium position and contribute to the damping. This partly accounts for the observed damping capacity of the base alloy. When the composite specimen is cooled, tensile thermal stresses arise in the matrix due to the particulate distribution and induce a long-range movement of the dislocations, which is superposed on the oscillatory motion imposed by the damping apparatus. This interaction is supposed to be the origin of the observed maxima. The dislocation density in the unreinforced matrix is quite low [24]. Hence, the damping capacity is suggested simply to come from the dynamic hysteresis of the lattice defects [25] such as the vacancy or interstitial due to the reinforcement. The stress-induced ordering process of these leads to an anelastic strain, and the damping mechanism is referred to as the linear reversible anelastic relaxation and it opens up the dynamic hysteresis loop. The dynamic hysteresis loop generally dissipates a smaller quantity of energy while the residual stresses are relaxed from the composite during the heating and cooling cycles [26].
The vibration frequency and strain amplitude dependence of damping maxima suggests that the relaxation processes are of thermal origin and are stress dependent. The expression for activation energy corresponding to the thermally activated relaxation process is given by Parrine & Schaller [21]: f 0 e(/k TP) =1/2 (6) Where f is the vibration frequency, 0 is the relaxation time, is the activation energy, k is the Boltzmann constant and TP is the peak temperature. The data used in evaluating the activation energies h and c corresponding to the heating and cooling peaks respectively of the 6% Al 2 O 3 reinforced composite is given in Fig. 4. The thermal stresses generated in the composite are given by: th =(h-c)/2v* (7) Where v* is the activation volume and is given by v* =400 b 3 (b is the length of Burgers vector of dislocations in the matrix alloy =0.25 nm). Calculated value of th =3.081 Mpa. V. CONCLUSION The thermal expansion and damping behaviour of Al6061 alloy matrix and composites with Al 2 O 3 as reinforcement has been investigated over a temperature range of 25C to 500C both in the heating and the cooling cycles. The thermal expansion studies and damping behaviour showed hysteresis residual strains and in addition to this, the damping behaviour showed maxima in the heating and cooling curves. This study revealed the presence of residual thermal stresses generated in the composite due to the difference in the CTE between the matrix and the reinforcement. The long-range mobility of dislocations in the metal matrix around the reinforcement seems to be the critical parameter for the stress relaxation at matrix/reinforcement interface. This parameter has been carefully characterized by damping measurements. The study of thermal stresses leading to plastic deformation in the matrix and the residual strains obtained is particularly useful in any high temperature applications of composites. The thermal stresses have been evaluated separately under thermal expansion studies and damping studies and the results in both the cases are found to be in good agreement with each other.
Figure 4: Data used in the Computation Activation Energies (eV)
ISSN 2277-5056 | 2012 Bonfring Bonfring International J ournal of Industrial Engineering and Management Science, Vol. 2, No. 4, December 2012 135
ISSN 2277-5056 | 2012 Bonfring for Ph.Ds. ACKNOWLEDGMENT The main author B.V. Madhu. is indebted to R. Panduranga Vittal (Principal, Govt. Polytechnic, Holenarasipura), Dr.M. Krishna (Professor and Director, R&D, R .V. College of Engineering, Bangalore) and all the staff members of Govt. Polytechnic, Holenarasipura for helping in making this paper. Last but not least he would also like to thank his wife and parents for supporting in each and every step while preparing this paper.
[20] Z.M. Sun, J .B. Li, Z.G. Wang and W.J . Li, Residual stresses in silicon carbide particulate reinforced aluminum composites, Acta Metall. Mater., Volume 40, No.11, Pp. 2961-2966, 1992. [21] N. Srikanth, D. Saravanaranganathan, M. Gupta, L. Lu and M.O. Lai, Modeling and determination of dynamic elastic modulus of magnesium based metal matrix composites, Mat. Sci & Tech., Volume 16, Pp. 309-314, 2000. [22] R.J . Perez, J . Zhang, M..N. Gungor and E.J . Lavernia, Damping behavior of 6061Al/Gr metal matrix composites, Metall. Transa., Volume 24A., Pp. 701-712, 1993. [23] L. Parrine and R. Schaller, Thermal Stresses in metal matrix composites studied by internal friction, Acta Mater., Volume 44, No. 12, Pp. 4881-4888, 1996. REFERENCES [24] Shanthashatry, M. Krishna and J ayagopal Uchil, A study on damping behaviour of aluminite particulate reinforced ZA-27 alloy metal matrix composites, J our. of Alloys and Compounds, Volume 314, Pp. 268- 274, 2001. [1] S.C. Sharma, B.M. Girish, R. Kamath and B.M. Satish, Fractography, fluidity, and tensile properties of aluminium/hematite particulate composites, J our. of Mat. Engg. & Perf., Volume 8, No. 3, Pp. 309-314, 1999. [25] Santhashastry, M. Krishna and J ayagopal Uchil, Effect of Thermal Stress on the Thermal Expansion and Damping Behaviour of ZA- 27/Aluminite Metal Matrix Composites, J our. of Mat. Engg. & Perf., Volume 10, No. 2, Pp. 220-224, 2001. [2] S.C. Sharma, K.H.W. Seah, B.M. Satish and B.M. Girish, Effect of short glass fibers on the mechanical properties of cast ZA-27 alloy composites, Mat. & Des., Volume 17, No. 5/6, Pp. 245-250, 1996. [26] L. Parrine and R. Schaller, Characterization of mechanical stresses in metal matrix composites by internal friction, Scripta Mater., Volume 28, Pp. 763-767, 1993. [3] P. Reynaud, Cyclic fatigue ceramic-matrix composites at ambient and elevated temperatures, Composite Science and Technology, Volume 56, Pp. 809-814, 1996.
[4] S.C. Tjong and Z.Y. Ma, The high-temperature creep behaviour of aluminium-matrix composite reinforced with SiC, Al 2 O 3 and TiB2 particles, Composite Science & Technology, Volume 57, Pp. 697-702, 1997.
B.V. Madhu, obtained B.E. in Mechanical Engg. in 2004 from VTU (Karnataka), M.Tech. in Industrial Automation and Robotics in 2006 from VTU (Karnataka), and pursuing Ph.D (Part Time) in Composite Materials from J awaharlal Nehru Technological University, Anantapur (AP). At present he is working as Lecturer of Mechanical Engg. Dept., Govt. Polytechnic, Holenarasipura, Karnataka. He has 5 years of teaching experience for both Engineering and Diploma students. His areas of interest are Composite Materials, MMC, and FEM. He is active member of ISTE. [5] H. Akbulut, M. Durman and F. Yilmaz, Higher temperature youngs modulus of aluminiumshort fibre reinforced Al-Si MMCs produced by liquid infiltration, Materials Science and Technology, Volume 14, Pp. 299-305, 1998. [6] S.C. Sharma, Effect of Al 2 O 3 particle on the coefficient of thermal expansion behavior of the Al6061 alloy composites, Metall. & Mat. Trans.A, Volume 31A, Pp. 773-780, 2000. [7] U. Rajendra, Vaidya and K.K. Chawla, Thermal expansion of metal- matrix composites, Comp. Sci. & Tech., Volume 50, Pp. 13-22, 1994.
[8] Eun U. Lee, Thermal stress and strain n a Metal Matrix Composites (MMCs) with a spherical reinforcement particle, Metall. Trans. A, Volume 23A, Pp. 2205-2210, 1992. Dr.K. Prahlada Rao obtained B.Tech. in Mechanical Engg. in 1982, M.Tech. in Refrigeration and Air Conditioning in 1985 from J NTU. At present he is working as Vice - Principal & Professor of Mechanical Engg. Dept., J NTUCEA, Anantapur (AP). He joined as Assistant Professor in Mechanical Engg. in Oct 1985, pursued Research Programme at University of Roorkee, Roorkee (1990-93), promoted as Associate Professor in 1996, pursued Ph.D in 2002 in Expert Systems for Design and Manufacture of Metal castings, promoted as Professor of Mechanical Engg in 2006, served as H.O.D of Mechanical Dept. (2008-10), published 80 papers in National and International journals/ conferences, and successfully guided 7 Research Scholars for Ph.Ds. His areas of research interest include: Artificial Intelligence / Expert Systems applications in Design / Production/ Industrial Engg. domains. He has served as Chairman- Board of Studies for Mechanical Engg .in 2009 at J NTU, Anantapur. [9] S.C. Sharma and A. Ramesh, Effect of Heat treatment of mechanical properties of particulate reinforced Al6061 composites, J our. of Mat. Engg. & Perf., Volume 9, No.5, Pp. 557-561, 2000. [10] K.H.W. Seah, S.C. Sharma and A. Ramesh, Mechanical properties of cast aluminiumalloy 6061- Al 2 O 3 particulate composites, Proc. Instn. Mech. Engrs., Volume 214(L), Pp. 1-6, 2000. [11] S.C. Sharma, K.H.W. Seah, M. Krishna and A. Ramesh, Oxidation behaviour of 6061 Al/ Al 2 O 3 composites at high temperature, Alloys and Compounds, Volume 306, Pp. 270-276, 2000. [12] S.C. Sharma, Oxidation behaviour of 6061 Al/ Al 2 O 3 composites at high temperature, J our. of Mat. Engg. & Perf., Volume 9, No. 3, Pp. 344-359, 2000. [13] S.C. Sharma, A study of wear mechanism in particle reinforced aluminiummetal matrix composites, Accepted for publication in J our. of Mat. Engg. & Perf.
[14] S.C. Sharma, A study on stress corrosion behavior of Al6061/ Al 2 O 3
composite in higher temperature acidic medium using autoclave, Corrosion Science, Volume 43, Pp. 1877-1889, 2001. Dr.D.P. Girish obtained B.E. in Mechanical Engg. In 1993 from Bangalore University, M.Tech. in Maintenance Engineering in 1997 from Mysore University, and Ph.D in 2007 from Bangalore University. At present he is working as Professor of Mechanical Engg. Dept., Govt. Engg. College, Ramanagara, Karnataka. He has 18 years of Teaching and 03 years of Industrial experience. His areas of interest are Composite Materials, MMC, FRP, and FEM. He has published 7 research papers in International J ournal and presented 18 papers in National and International Conferences. He is active member of ISTE, FIE, and CE. He is successfully guiding 5 Research Scholars [15] H.M. Ledbetter and M.W. Austin, Internal strain (stress) in a SiC-Al particle reinforced composite: an X-ray diffraction study, Mater. Sci. &Eng., Volume 89, Pp. 53-61, 1987. [16] E. Carreno-Morelli, S.E. Urreta and R. Schaller, Mechanical spectroscopy of thermal stress relaxation at metal-ceramic interfaces in aluminium-based composites, Acta Mater., Volume 48, Pp. 4725-4733, 2000. [17] S.C. Sharma, The effect of ageing duration on mechanical properties of al 6061-garnet composites, Proc. Instn. Mech. Engr, Volume 215(L), Pp. 113-119, 2001. [18] C.T. Kim, J .K. Lee, and M.R. Plichta, Plastic relaxation of thermoelastic stress in aluminium/ceramic composites, Metall. Trans. A, Volume 21A, Pp. 673-682, 1990. [19] S. Berger, Elastic and plastic strains in Al/TiW/Si contacts during thermal cycles, Mat. Sci. & Engg., Volume A288, Pp. 164-167, 2000.