JOURNAL DE PHYSIQUE Colloque C8, supplgment au n012, Tome 48, d6cembx-e 1987 INTERNAL FRICTION AND DYNAMIC MODULUS OF METAL MATRIX COMPOSITES AND ADVANCED ALLOYS A. WOLFENDEN, C.K. FRISBY, K.J. HERITAGE, S.S. VINSON* and R.C. KNIGHT* Mechanical Engineering Department, Texas A & M U n i v e r s i t y , College S t a t i o n , TX 77843, U . S . A . *LTV Missiles and E l e c t r o n i c s Group, Missiles D i v i s i o n , PO Box 650003, D a l l a s , TX 75265, U . S . A . ABSTRACT The PUCOT (piezoelectric ultrasonic composite oscillator technique) has been used at frequencies near 100 kHz to measure internal friction and dynamic Young's modulus of various metal matrix composites (MMCs) and advanced alloys. The materials in the study were: Al/SiC MMCs with up to 20 volume % Sic and powder metallurgy (PM) A1-Fe-X alloys denoted as 452 and B014L. The testing was performed at various temperatures ranging from room temperature to over 300°C. The strain arnplhtude dependence of internal friction was investigated over the strain range 10- to The modulus data were fitted to a linear equation of the type: E = E 0 - MT, where () () E (GPa) is the dynamic modulus at temperature T (OC), E 0 is the modulus at O°C and M is the slope dE/dT in GPa/OC. The values of M for the materials studied varied in P/? the range 0.03 to 0.10 $ a ' , while values of M/E(O) (= -(l/E)(dE/dT)) fell in the interval (4 to 9 x 10- OC- ) . The effects of a flash anneal (540°C for 5 minutes) on the dynamic modulus (measured at room temperature) for the PM (Powder Metallurgy) Aluminum specimens was also investigated. The PUCOT is described, and the damping and dynamic modulus data are discussed. INTRODUCTION Aluminum matrix composites and powder metallurgy materials are major research areas in the aerospace industry. This field is particularly interested in materials that have a large Young's modulus to density ratio (specific modulus). Powder metallurgy (PM) aluminums and metal matrix composites (MMCs) have been supplied by LTV Aerospace & Defense Co. for testing. Several manufacturing processes have been varied, such as: type of matrix aluminum, rolling direction of PM alloy and volume fraction of Sic. These materials must withstand elevated service temperatures without an appreciable loss in mechanical properties. All of these variables must be considered so that an optimum material can be chosen for a specific application. This paper reports the results of an investigation using the PUCOT (piezoelectric ultrasonic composite oscillator technique) to determine internal friction and dynamic Young's modulus at temperatures up to slightly above 300°C. EXPERIMENTAL PROCEDURES The PUCOT (1-4) was used to determine internal friction and dynamic Young's modulus at room temperature and at elevated temperatures. Since the PUCOT technique is not well-known, a brief description of the apparatus and its operation is given here. A schematic of the PUCOT is shown in Figure 1. The technique incorporates two identical alpha-quartz piezoelectric crystals (driver and gage), a spacer rod (used at elevated temperatures), and a test specimen. The two crystals are joined together followed by the addition of the quartz spacer rod using Loctite glue. Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1987856 C8-378 JOURNAL DE PHYSIQUE Since the specimen is to sustain high temperature, it is cemented to the end of the spacer rod using ceramic cement. At ambient temperatures, the spacer rod is omitted and the test specimen is glued directly to the crystals as seen in the three . component system of Figure 1 In this arrangement, the piezoelectric crystals excite longitudinal ultrasonic stress waves in the test specimen of appropriate resonant length. The resonant system is driven by a closed-loop oscillator that maintains a constant maximum strain amplitude in the test specimen. During the experiment the resonant period, the drive voltage, gage voltage and temperature (above ambient temperature) are measured. Other recorded values are the density, mass and length of the test specimen along with the mass and period of the piezoelectric crystals and spacer rod. The thermal expansion coefficient is used for corrections in length and density at elevated temperatures. Since the quartz rod is tuned to resonate at a particular temperature, measurements of damping and modulus are made at specific, selected temperatures only, in our case 20, 100, 175, 200 and 316OC (depending on the alloy). The data analysis was completed on a Commodore 64 computer equipped with appropriate programs. MATERIALS Metal Matrix Composites: Three categories of cast A 357Al/SiC composites were investigated. Aluminum A 357 is a castable aluminum alloy with a Young's modulus of approximately 71 GPa depending on heat treatment. The MMCs had various volume percent of silicon carbide whiskers (10 v/o, 15 v/o, 20 v/o). Each group consisted of several test specimens. The cast specimens measured 5 mm x 5 mm x 40 mm. Powder Metallurgy Alumi~ums: Alloys 452 and B014L are advanced powder metallurgy aluminum alloys that are produced using rapid solidification processing techniques. The alloys are produced by Allied-Signal Corporation, the composition of the individual alloys being tailored to produce desired properties. Alloy 452 is a high stiffness composition that exhibits exceptional elevated temperature properties. Alloy B014L is designed to exhibit slightly lower stiffness and strength properties, but high fracture toughness and high corrosion resistance characteristics. The novel (and proprietary) alloy compositions give rise to thermally stable, fine dispersoids that improve properties over those of conventional aluminums. RESULTS AND DISCUSSION A 357Al/SiC: For these specimens modulus was measured at all temperatures and damping was found at 100 and 200°C. The av ra e values of damping and modulus for the 10 v/o 2 specimens at41000C were x loag atd 82.5 GPa, respectively. The damping ranged from 4 x 10- to 6 x 10- and the modulus ranged from 81.2 to 83.4 GPa. The temperature dependence of the modulus was fitted to a linear equation of the form: E = 88.4 - 0.05T (GPa, OC); the correlation coefficient averaged 0.99 for the 10 v/o specimens. The 15 r/o specimens t 100°C showed an average damping of 8.8 x ) (ranging from 4 x 10- to 1.5 x 10- and an average modulus of 87.1 GPa (ranging from 85.2 to 89.7 GPa). The linear equation for the temperature dependence of modulus was E = 87.2 - 0.09T and the correlation coefficient averaged 0.96. Fina ly, for the 20 v/o spe~imensthe avesage damping and modulus at 100°C were o' 1.4 x l- (ranging from 8 x 10- to 1.8 x 10- ) and 89.9 GPa (ranging from 89.6 to 90.3 GPa), respectively. The linear equation for the temperature dependence of modulus was E = 95.9 - 0.05T and the correlation coefficient averaged 0.97 for the 20 v/o specimens. PM Al: Two types of PM aluminum specimens were provided: PM 452 and PM B014L. Each type was further divided into two orientations, perpendicular and parallel. The parallel orientation refers to the specimen being cut parallel to the rolling direction, while the perpendicular orientation refers to a perpendicular cut. The damping and modulus values were found at room temperature (20°C) and at an elevated temperature of 300°C. These same properties were also found after a flash anneal of 540°C for 5 minutes. The PM 452 specimens had an average damping value of at ro m temperature for both orientations. The values ranged from 8 x to 1.4 x 10-3 . The dynamic modulus ranged from 91.3 GPa to 95.6 GPa with an average of 94.4 GPa. The amping values at room temperature for the PM B0145 specimens ranged from 1.1 x l- to 1.9 x o' with an average value of 1.4 x 10-. The modulus averaged 86.7 GPa and did not deviate a significant amount from this value. The damping and dynamic modulus were again found at room temperature after the exposure of the specimens to ax elevated tem erature of 540°C for five m'nutes. The o' damping ranged between 5 x 10- to 1.7 x l - with an average of 9 x 10f for the PM 452 specimens. The modulus for these specimens averaged 95.4 Pa. The damping valyes of the PM B014L specimens fell between 6 x 10- and (average = 8 x 10- ), while the modulus ranged between 85.0 GPa and 86.6 GPa (av5rage = 85.6 GPa). At 316°C 5he average values of damping and modulus were 4.7 x 10- and 82.4 GPa, and 4.9 x 10- and 73.0 GPa for the 452 and B014L materials, respectively. The source of internal friction in MMCs is an intriguing topic. An examination of the data for Al/SiC in Table 1 reveals that3 at 100°C the damping increases by a factor of three (from 4.7 x to 1.43 x 10- , average values) as the volume percentage of Sic increases from 10 to 20%. At 200°C, however, 3 partial reve se trend is jndicated: the internal friction ranges from 2.2 x 10- to 7.7 x l- to o' 1.2 x 10- as the volume percentage of Sic goes from 10 to 15 to 20%. A possible interpretation of these results is suggested by the TEM observations of Fisher and Arsenault (5) of high concentrations of dislocations near the Al/SiC interfaces. The authors proposed that thermal stresses at the interfaces were induced on cooling the MMC and these stresses generated large numbers of dislocations. Thus, in our specimens we postulate that for the test temperature of 100°C such high dislocation concentrations may exist, leading to higher damping levels as the percentagae of Sic is increased, while at 200°C, where significant annealing out of dislocations in the Al/SiC ought to occur, the damping level due to the dislocations near the Al/SiC interfaces should decrease. In support of this postulate, we may note that for the PM A1 alloys, which have been formulated to contain thermally stable, fine dispersoids (to prevent significant annealing out of dislocations at higher temperatures), the internal friction increases by a factaor of 4.5 (for PM 452) or 3.3 (for PM B014L) as the test temperature increases from 29 to 316°C. In other words, the PM A1 alloys behave in a conventional manner (Q- increases as temperature increases), whereas the MMCs do not, due to the unusually high dislocation concentrations near the Al/SiC interface. Thus, it apears that a significant contribution to internal friction over a certain temperature range in MMCs may arise from the high concentration of dislocations near the fiberhatrix interfaces. CONCLUSIONS Many conclusions can be drawn from the data obtained. First of all, the modulus and damping values of the PM aluminum specimens are not significantly dependent on the rolling direction. The properties of the specimens remained fairly constant at a given temperature. As expected, the damping increased by a factor of about five and the modulus decreased by about 12 GPa with the increase in temperature of approximately 300°C. The flash anneal had little effect on the damping or the modulus. A longer exposure time would probably have resulted in more significant changes in the material properties. It was noted from the average modulus values found for the Al/SiC composites for each volume percent category that the 20 volume percent material has the highest stiffness properties at all temperatures. For all the mat rial tested the values of M/E(O) (=-(l/E)(dE/dT)) fell in the range (4 to 9) x l- ' - . This agrees with the range noted by Friedel (6) for o' CI many elements. It seems likely that significant internal friction over a certain temperature range in MMCs may arise from the high concentrations of dislocations near the fiber/matrix interfaces. C8-380 J O U R N A L D E PHYSIQUE TABLE 1: ALUMINUM/SILICON CARBIDE DATA Spécimen Temp. Q"1 Modulus Temp. Q"1 Modulus Temp. Q"1 Modulus ID -3 -3 3 <°C) (xlO ) (GPa) (°C) (xlO ) (GPa) <°C) ( x H F ) (GPa) SIC A357 10%#1 20 — 87.1 100 — — 200 — — 102#2 20 — 87.5 100 0.4 81.2 200 1.5 77.7 10£#3 20 — 88.8 100 0.6 83.4 200 3.1 78.2 102#4 20 — 87.5 100 0.4 82.9 200 2.0 78.5 15X#1 100 0.7 87.4 175 200 0.7 88.6 152#2 100 1.0 89.7 175 1.0 89.5 200 0.7 84.8 15X#3 100 0.4 86.9 175 1.3 85.0 200 — — 15£#4 100 0.8 86.2 175 1.3 83.6 200 — — 152#5 100 1.5 85.2 175 1.5 87.2 200 0.9 84.0 20Z#1 20 95.6 100 200 202#2 20 — 95.8 100 1.7 89.6 200 1.3 86.8 20X#3 20 — 95.0 100 1.8 90.3 200 0.8 86.2 20X14 20 — 95.1 100 0.8 89.7 200 1.5 85.7 TABLE 2 : PM ALUMINUM DATA Spécimen Temp. Q"1 Modulus Q"1* Modulus* Temp. Q"1 Modulus -3 -3 -3 (°C) (xlO ) (GPa) (xlO ) (GPa) (°C) (xlO ) (GPa) PM 452 #1 PAR 20 0.9 95.1 1.7 94.2 316 2.9 85.8 #2 PAR 20 1.4 95.6 0.6 95.6 316 7.7 78.8 #1 PER 20 0.8 91.3 0.5 96.7 316 4.4 85.0 #2 PER 20 1.0 95.6 0.6 95.0 316 3.6 80.0 PM B014L #1 PAR 20 1.1 86.7 0.7 85.0 316 4.7 71.6 #2 PAR 20 1.1 86.7 0.6 85.0 316 3.5 74.4 #1 PER 20 1.8 86.7 1.0 86.6 316 7.5 73.0 #2 PER 20 1.9 86.6 1.0 85.8 316 3.9 72.9 PAR: Parallel to rplling direction PER: Perpendicular to rolling direction *After flash anneal at 540°C for 5 minutes. REFERENCES 1. J. Marx, Rev. Sci. Instr. 22, 503-509 ( 1 9 5 1 ) . 2. W.H. Robinson and A. Edgar, Trans. Sonics and Ultrasonics, IEEE, E,98-105 (1974). 3. J.L. Tallon and A . Uolfenden, J. Phys. and Chem. of Solids, 40, 831-837 ( 1 9 7 9 ) . 4. M.R. Harmouche and A. Wolfenden, J. of Testing and Evaluation, 13,424-428 (1986). 5. R.J. Arsenault and R.M. Fisher, Scripta Met. 17,67-71 (1983). 6. J. Friedel, "Dislocations", Pergammon Press, New York, 1964, Appendix B, 454-457. SCHEMATIC VIEW OF PUCOT SPECIMEN ASSEMBLY THREE COMPONENT FOUR COMPONENT SYSTEM SYSTEM SPECIMEN 1 fli SPACER ROD I I, I 1, SPECIMEN F""A"E Figure 1 : Schematic drawing of the PUCOT. Left hand side: apparatus for measurements at room temperature; right hand side: for measurements at elevated temperatures.
Pages to are hidden for
"Dallas TX"Please download to view full document