IV1£3I_J RETURNING MATERIALS: Place in book drop to ”Bumgs remove this checkout from -_ your record. FINES win be charged if book is returned after the date stamped below. m 2‘s 1995 EFFECT OF THERMOMECHANICAL TREATMENTS ON THE TRANSFORMATION BEHAVIOR OF A TiNi ALLOY BY Kuang-Hua Hou A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Metallurgy, Mechanics and Materials Science 1986 YoS/l 73? ABSTRACT EFFECT OF THERMOMECHANICAL TREATMENT ON TRANSFORMATION BEHAVIOR OF A TiNi ALLOY BY Kuang-Hua Hou The transformation behavior of a TiNi alloy has been studied by using electrical resistance vs. temperature curves obtained from different thermomechanically treated specimens. Transmission electron microscopy and electron diffraction were also employed to observed the microstructures of specimens. Effects of aging time, aging temperature, cooling rate, thermal cycling, cold rolling and annealing after cold rolling were investigated. By combining the effects of different variables, the controlling factors of transformation behavior of this alloy have been determined to be the formation of precipitates and the introduction of dislocations. The former affects the composition of the matrix and the latter affects the formation of martensite plates. Premartensitic transformation has been investigated by TEM. Charge density wave influences the structure of parent 82 phase by changing it to incommensurate or commensurate phases. Premartensitic transformation can be enhanced by aging treatments. TABLE OF CONTENTS page ACKNOWLEDGEMENTS .... ................................. iii LIST OF TABLES ....... ................................ iv LIST OF FIGURES ...................................... V I. INTRODUCTION ........... ..... ... ..... . ........... 1 II. EXPERIMENTAL PROCEDURES ......................... 10 III. RESULTS AND DISCUSSION .......................... 13 A) As-quenched specimen ..... ..................... 13 B) Effect of aging temperature ..... .............. 21 C) Effect of aging time .......... ................ 36 D) Effect of cooling rate .. ...................... 50 E) Effect of thermal cycling ..... ................ 54 F) Effect of cold rolling ............ ............ 60 G) Effect of annealing after cold rolling ........ 68 IV. SUMMARY........... ...... ................ .......... 78 LIST OF REFERENCES ........................... ........ 81 ii The Dr. C. advices The ACKNOWLEDGEMENTS author wishes to express sincere appreciation to M. Hwang for his patient guidance and encouraging during this research. help from my friend Pei-San Yu is very much appreciated. This thesis is to my dearest parents. iii LIST OF TABLES Table page 1. Transition temperatures of annealed and aged SpeCimens O.......OOOOOOOOOOOOOOOOO ....... l4 2. Transition temperatures of cold-rolled and "cold-rolled and annealed" specimens .. ........ 15 3. Thermal cycling effect on transition temperatures ......OOOOOOOOOOOOOOOO ............ 59 iv Figure 1(a). 1(b). 2(a). 2(b). 2(c). 2(a). 6(a). 6(b). 6(c). 6(d). LIST OF FIGURES page Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196 C, As-quenched specimen..... ......... 16 Electrical resistance vs. temperature curve of thermal cycle 90 between +60°C and -196°C, As-quenched specimen .............. 17 Bright field micrograph of As-quenched specimen, mag. 50,000X ....... ................ 19 [111182 zone diffraction pattern of As- quenched specimen .................. .......... 19 [110182 zone diffraction pattern of As- quenched specimen ...................... ...... 20 [210]BZ zone diffraction pattern of As- quenched specimen ..................... ....... 20 Electrical resistance vs. temperature curve of thermal cycle 1 between +6d’C and -196°C, specimen sod'C-lhr-water . ........ 23 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196’C, specimen 700°C-1hr-water . ........ 24 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen 6od’C-1hr-water . ........ 25 Electrical resistance vs. temperature curve of thermal cycle 1 between +60’C and -196°C, specimen 500°C-1hr-water . ........ 26 Bright field micrograph of specimen 500°C—1hr-water, mag. 300,000x .... ........... 27 Dark field micrograph of specimen 500°C-1hr-water, mag. 300,000X ..... .......... 27 Diffraction pattern corresponding to 6(c) and 6(d) ................ ............. 28 Figure 7(a). 7(b). 7(c). 7(d). 7(e). 7(f). 10. 11(a). 11(b) . 11(c). 11(d) . 11(e). page Electrical resistance vs. temperature curve of thermal cycle 1 between +60 C and ~196°C, specimen 400°C-1hr-water . ........ 30 Electrical resistance vs. temperature curve of thermal cycle 90 between +60 C and -l96°C, specimen 4od'C-1hr-water ......... 31 Electron diffraction pattern of specimen 40d’C-1hr-water .................... .......... 32 Dark field micrograph of specimen 400°C-1hr-water, mag. 300,000X ....... ........ 32 Bright field micrograph corresponding to 7(d) 0000000000000000000000000... 0000000000 33 Selected area diffraction pattern corresponding to 7(d) abd 7(e) ..... .......... 33 Electrical resistance vs. temperature curve of thermal cycle 1 between +60 C and -196°C, specimen 300°C-1hr-water .. ....... 34 Aging temperature effect on transition temperatures 000.00.000.0000000000.00...0.00000 35 Electrical resistance vs. temperature curve of thermal cycle 1 between +60 and -196°C, specimen 500 C-4hrs-water ..... ....... 38 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen 500 C-10hrs-water ....... 39 Bright field micrograph of specimen 500°C-10hrs-water, mag. 150,000X .... ......... 40 Dark field micrograph of specimen SOO‘C-thrs-water, mag. 200,000X ..... ........ 40 Dark field micrograph of specimen 500°C-10hrs-water corresponding to 11(C), mag.200'000x 00.000.000.00... 000000000 41 Selected area diffraction pattern corresponding to 11(c) and 11(d) .... ......... 41 vi Figure 12. 13. 14(a). 14(b) . 14(c). 14 (d) . 14(a). 14(f). 14(9)- 14 (h) . 15(a). 15(b) . 15(c). page Aging time effect on transition temperatures of specimens aged at 500°C00.00000000000000000000000000 00000000000 42 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen 400 C-4hrs-water ........ 43 Electrical resistance vs.temperature curve of thermal cycle 1 between +60°C and -196 C, specimen 400'C-10hrs-water ....... 45 Electrical resistance vs. temperature curve of thermal cycle 90, between +60°C and -196'C,specimen 400° C-thrs-water . ...... 46 [1101B2 zone diffraction pattern with 1/3 <101> extra spots, specimen 400 C-10hrs- water 0000.00.000.000000000000000000000 0000000 47 [1101B2 zone diffraction pattern with 1/3 <111> extra spots, specimen 400 C-thrs- water 00000000.00.000.0000000000000000 00000000 47 Bright field micrograph corresponding to 14(d)’ mag. SOIOOOX 00000000000000.0000 0000000 48 [111132 zone diffraction pattern with 1/3 <011> extra spots, specimen 400 C-10hrs- water 000000000000000000000000.000000000 000000 48 Dark field micrograph of specimen 400 C— 10hrs-water, mag. 100,000X ............. ...... 49 Bright field micrograph corresponding to 14(g), mag.100,000X ................... ....... 49 Electrical resistance vs. temperature curve of thermal cycle 1 between +60‘C and -196’C, specimen 900°C-1hr-air ... ...... .. 52 [210132 zone diffraction pattern with diffuse streaks, specimen 900°C-1hr-air ...... 53 Bright field micrograph of specimen 900°C-1hr-air, mag. 60,000X .......... ........ 53 vii Figure 16(a). 16(b). 16(c). 16(d). 16(e). 17. 18(a). 18(b). 18(c). 18(d). 18(e). 18(f). 18(g). 19. page Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen 400°C-1hr-air .. ......... 55 [111132 zone diffraction pattern of specimen 400’C-1hr-air .......... ............. 56 Bright field micrograph of specimen 400’C-1hr-air, mag. 150,ooox ..... ............ 56 Bright field micrograph of specimen 400°C-1hr-air, mag. 50,000X ...... ............ 57 Selected area diffraction pattern corresponding to 16(d) ............. .......... 57 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen C-R-10% .......... ....... 62 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen C-R—20% ................. 63 Bright field micrograph of specimen C-R-20%’ mag. 30'OOOX 00.000000000000000000000 64 Dark field micrograph of specimen C-R-20%' mag. SOIOOOX 00000000000000... 00000 00 64 Bright field micrograph corresponding to 18(c), mag. 50,000X ............... ........ 65 Selected area diffraction pattern corresponding to 18(c) and 18 (d) .. .......... 65 [110132 zone diffraction pattern with 1/3<112> extra spots .......... .......... 66 [111132 zone diffraction pattern with diffuse streaks ................. ........ 66 Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°c, specimen C-R-40% ........... ...... 67 viii Figure 20. 21(a). 21(b). 21(c). 21(a). 21(e). 21(f). 21(g). 21(h). 22. page Electrical resistance vs. temperature curve of thermal cycle 1 between +60’ and -196‘C, specimen C-R-10%--500'C- 1hr-water ......................... ........... Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°c, specimen c-R-zo%—-soo°c- 1hr-water ......................... ........... Bright field micrograph of specimen C-R-20%--500’C-lhr-water, mag. 12,ooox ....... Bright fieldomicrograph of specimen C-R-20%--500 C-lhr-water, mag. 12,000X ....... Dark field micrograph of specimen C-R-20%--500°C-lhr-water, mag. 100,000x ....... Bright field micrograph corresponding to 21(d), mag. 100,000x ...................... Selected area diffraction corresponding to 21(d) and 21(8) 00000000000000.0000. 0000000 Bright field micrograph of specimen C-R-20%--500°C-lhr-water, mag. 80,000X ....... Enlarged bright field micrograph from 21(g), mag. 300,000X .................... Electrical resistance vs. temperature curve of thermal cycle 1 between +60°C and -196°C, specimen C-R-40%--500°C- lhr-water .................................... ix 7O 71 72 73 73 74 74 75 76 I. INTRODUCTION The binary TiNi alloys have been known as shape memory alloys since their unusual shape memory effect (SME) was reported in 1962 by William J. Buehler of the' U.S. Naval Ordance Laboratory. Extensive studies of TiNi alloys have been done in order to understand the mechanisms of SME and its related transformations. Besides SME, pseudoelasticity and "two way" shape memory effect are also studied as important characteristics which make TiNi alloys as the most promising alloys in industrial as well as medical applications (1-5). The high temperature parent phase of TiNi has already been determined as having CsCl (32) structure (6-15). As TiNi is cooled from parent phase to just above Martensite start temperature (Ms), it will exhibit some anomalies in physical properties, such as rising electrical resistance, decreasing sound velocity, specific heat peaks, internal friction peaks, etc. These phenomena may be attributed to the premartensitic transformation (6-15). The premartensitic transformation of TiNi is suspected to be associated with a lattice modulation, due to periodic displacement of atoms. This lattice modulation can lower the energy of the materials and probably is electronic in origin. The formation of a charge density wave (CDW) was proposed to explain the premartensitic transformation behaviors (6-12). "A charge density wave is a static modulation of the conduction electrons, which is a Fermi- surface-driven phenomenon usually accompanied by a periodic lattice distortion." (6) " A favorable Fermi surface geometry is required for the formation of a CDW. A CDW will most likely occur when the shape of the Fermi surface can be connected by the same wave-vector Q., i.e. Q = 2Kf."(6) The presence of the CDW will alter the normal crystalline periodicity of the material because ion displacements appear to stabilize the charge perturbation. Because of the unlikelhood of favorable Fermi surface nesting, charge density waves are restricted in one- or two- dimensional materials. Though CDW phenomena are rarely found in three- dimensional materials, it is possible to find the formation of CDWs in the transition metals, due to their complex overlapping of d and f bands (6-12). The wave vector of a CDW is not necessarily an integral fraction of a reciprocal lattice vector of the undistorted parent phase. The material exhibiting this phenomenon is in a quasi-crystalline phase called "incommensurate phase" (6). A so-called "commensurate" or "locked-in" phase is the state when the wave vector of a CDW is exactly an integral fraction of its parent phase (7). Therefore, the formation of a CDW can be accompanied by two structural phase transformations which are the formations of incommensurate and commensurate phases. The superlattice resulting from the formation of a CDW will create some satellite diffractions near the Bragg diffractions of the parent lattice (6-17). In TiNi alloys, the 1/3 position reflections are the results of the superlattice arising from a CDW. When TiNi is cooled from its high temperature 32 parent phase, the first premartensitic transformation, which is second order, will start at Tp temperature, where the electrical resistance begins to increase and the structure has changed to distorted 32 incommensurate phase. Some diffuse streaks and extra diffraction spots can be found among the 32 diffraction pattern. The accompanying microstructural change gave rise to the separate antiphase domains can be found in the parent matrix (6-14). With further cooling the incommensurate phase will change to commensurate phase at Td temperature, which is at the reflection point of the resistance curve between Tp and martensite start (Ms) temperatures. This is called the second premartensitic transformation which is first order. The structure will change from distorted 32 to rhombohedral R phase, while extra diffraction spots appear at exact 1/3 positions of rhombohedral diffraction pattern. At this stage , needle-like domains can be found associated with the second premartensitic transformation in the matrix. Martensitic transformation starts with the abrupt decrease of electrical resistance due to the formation of martensite plates. R phases can coexist with martensite plates until Mf, martensite finish, temperature is reached. The crystal structure of martensite plates is distorted monoclinic 319'. The sequence of martensitic transformation can be described as follows : parent phase (32)--- incommensurate phase (distorted cubic)--- commensurate phase (rhombohedral)---martensitic phase (monoclinic, 319') (6-8, 13,14,17). Upon heating, austenite starts to form at As temperature, which is associated with an abrupt increase of electrical resistance. As the resistance begins to decrease, the austenite finish temperature (Af) is achieved. At this time, the structure has returned to its high temperature parent phase. There are several factors affecting the martensitic transformation behavior of TiNi alloys, including alloy composition, thermal cycling, aging temperature, aging time, cooling rate, cold work, annealing after cold work, etc. The transformation behaviors of TiNi alloys are very sensitive to the relative concentration of Ti and Ni (10,18). The more Ni in the alloy, the lower the Ms temperatures is. The precipitate styles formed during aging treatments are controlled by composition of alloys (18). At least two types of precipitate have been found, i.e. TiZNi and TiNi3. The former appears in Ti rich alloys and the latter in Ni rich alloys. The formation of precipitates will change the matrix composition, which could raise or lower the Ms temperature, according to which type of precipitate is formed. The alloy used in this study is Ni rich; therefore, the formation of precipitate TiNi3 during aging treatments will decrease the concentration of Ni in the matrix. Thus, Ms temperature becomes higher. Alloying elements such as Cr, Cu, Fe and Al can trap vacancies in the alloy and therefore lower the Ms temperature (6-8, 10). The effects of composition and alloying elements are not studied in this research. The increase of thermal cycling can suppress the Ms temperature to a different extent according to the thermomechanical history of the different specimens (19,20). In this study, the As-quenched specimen had a 100 C difference in Ms temperature after 90 thermal cycles from +6d’C to -196°C, while 400°C-10hrs-water had only 2° C during the same thermal cycling. The dislocations introduced during thermal cycling impede the martensite formation, consequently depressing the Ms temperature. The effect arising from dislocations can be partially or almost totally eliminated by the formation of precipitates, which may obstruct the mobility of dislocations. The effects of aging temperature can be separated into two groups, i.e. higher temperature aging and lower temperature aging. In this study, if the aging temperatures are higher than or equal to 606’ C, the resulted electrical resistances will be very similar to that the of As-quenched specimen. The effect of aging at this temperature range is to shift the electrical hystereses towards higher temperature. When the specimens are aged at 500°C or lower, the formation of precipitates will lower the Ni concentration in the matrix, thus raising the Ms temperatures. If the size of the precipitates is too small, the coherency of precipitates and matrix will obstruct the martensite formation. This was the case with the specimen aged at 300°C for 1 hour and quenched into water. The effect of aging time is evaluated by aging specimens at 406’C and 500°C for 1 hour to 10 hours. When aging time extends from 1 to 4 hours, the precipitates in the matrix will grow larger and consume more Ni from the matrix. This phenomenon will raise the Ms temperature and change the shape of the electrical resistance curve. Further aging from 4 to 10 hours has no apparent evident effect on either transition temperatures or shape of the electrical resistance curve. The equilibrium condition created between precipitates and matrix prohibits the further growth of precipitates. Therefore, the effects of aging for longer time periods are not obvious. The effects of cooling rate on TiNi alloys may be divided into two groups. When the specimen is aged at higher temperatures, the slow cooling rate will introduce the opportunity of lower temperature aging. Though this lower temperature aging time is very short, a certain amount of precipitate can form. Since the aging time period is only seconds long, the precipitates aggregate along the grain boundaries in order to lower their nucleation energy. This phenomenon could probably be responsible for the depression of electrical resistance hysteresis. When aged at lower temperatures, precipitates have already formed in the matrix and there is no difference in transition temperatures and electrical resistance curves between air-cooled and water-quenched specimens. After the specimen was cold rolled, dislocations were introduced. The Ms temperature is depressed by the large quantity of dislocations which partially or completely prevent the stable reorientation or the growth of the martensite plates. These dislocations also affect the mobility of other dislocations which are responsible for the plastic flow. Many strain-induced martensite plates are formed in the matrix (16,17 19,21-23). These martensite plates are not thought to be mobile because of dense dislocations around them. In TEM observations, 32 diffraction pattern with diffuse streaks, extra 1/3 position diffraction spots and diffraction spots arising from martensite plates can be found in the same specimen. This represents the coexistence of parent phase, R phase and martensite phase. It is suspected that the R phases could probably nucleate preferentially on dislocations to lower their nucleation energy. The effect of annealing on cold-rolled specimens is enormous (24), no matter what the degree of mechanical deformation. Annealing will dramatically change the transition temperatures and the shape of electrical resistance vs. temperature curves by creating a dislocation substructure. This substructure can obstruct the growth of martensite plates, but makes the reorientation of these plates easier. This may lead to the formation of microtwins, which are suspected of being responsible for the pseudoelasticity below Ms temperature. Since the surprising similarity of transition temperatures and electrical resistance vs. temperature curves between "cold-rolled and annealed" specimens and specimens aged at same temperature, it may be concluded that the effect of the formation of precipitates is larger than that of the introduction of dislocations. In this study, all the factors described above, except alloy composition, are combined to investigate the effect of thermomechanical treatments on the transformation behaviors of a TiNi alloy. It is concluded that the formation of precipitates and the introduction of dislocations are two controlling factors affecting martensitic, as well as premartensitic, transformations. II. EXPERIMENTAL PROCEDURES A Ti-51.7Ni (at. pct.) alloy was used for this study. The bulk material was a 2cm x 2cm x 10cm bar, which was cut into 50.0mm x 2.0mm x 0.3mm strips for electrical resistance tests. The strips were sealed in quartz tubes under vacuum condition and were annealed at 900°C for 1 hour followed by quenching into water by breaking the tubes. These annealed specimens were denoted by "As- quenched" for convenience. In this study, all the annealing and aging heat treatments were performed while the specimens were sealed in quartz tubes under vacuum condition. The aged specimens were aged at different temperatures, ranging from 300°C to 900°C for 1 hr to 10 hrs. The effect of cooling rate on TiNi was evaluated by observing the two different rates obtained by water quenching and air cooling, after specimens had been aged. The aged specimens were denoted by their aging temperatures, aging time and cooling rate such as 900°C- lhr-air and 400°C-10hrs-water, etc (See Table 1). The cold-rolled specimens for electrical resistance tests were cut into 0.5mm thick strips, following the same heat treatment as that for As-quenched specimens. These 0.5 mm thick strips were rolled at ambient temperature to 40%, 20% and 10% reduction in thickness, and they are denoted as 10 11 C-R-40%, C-R-20% and C-R-10%, respectively. These cold— rolled specimens were then mechanically polished to standard size, i.e., 50.0mm x 2.0mm x 0.3mm. The "cold- rolled and annealed" specimens were made following the same procedures as cold-rolled specimens and then were annealed at 500° C for 1 hour and quenched into water. These specimens, corresponding to 40%, 20% and 10% reduction in thickness, were denoted by C-R-40%--Sod’C-lhr-water, C-R- 20%--soo° C-1hr-water and C-R-lO%--500° C-lhr-water, respectively (See Table 2). Before electrical resistance tests, all the specimens were electropolished in electrolyte consisting of 8% perchloric acid and 92% glacial acetic acid ( by volume ) by applying 25 volts in order to get rid of the oxidized layer on sample surface caused by heat treatments. The electrical resistance tests were caried out from +6d’C to -196‘C by using an Omnigraphic 200 X-Y recorder. The electrical resistance vs. temperature curves were automatically recorded. The controlling of temperature while cooling was performed by immersing the specimens into or withdrawing them out of a deep, wide-mouth container holding liquid nitrogen. A rate of 3.3cm/minute was maintained using a Hurst stepping motor. For heating, the specimens were put in an 11 ohm resistor coil set in a stable air condition and 30 volts in direct current were applied with the stepping motor shut off. 12 A copper-constantan thermocouple was spot-welded to the centers of the specimens while four copper wires were also spot- welded to the ends. The outer two copper wires were connected to the d.c. power supply which was a current source, and the inner two were connected to the X-Y 200 recorder. The d.c. power supply applied 140 V to the specimens, with a 1000 ohm resistor producing a current of 140 mA. The TEM specimens were cut from the bulk material into strips with thickness 0.25mm and 0.5mm for non-rolled and rolled specimens respectively. The same thermomechnical treatments as performed on the corresponding electrical resistance specimens were performed on these strips. They were then mechanically polished and cut into small discs, with 3mm in diameter and 0.15mm thick. The reason for using thick strips to prepare TEM specimens was to avoid surface artifacts probably caused by contamination from oxygen and/or nitrogen diffusion during heat treatments (25). Jet polishing was carried out at room temperature in a Struers Tenupol II by using 8% perchloric acid and 92% glacial acetic acid ( by volume ) as electrolyte while applying 20 volts d.c. The TEM observations were conducted in a Hitachi 'H-800 Transmission Electron Microscope operating at 200 KV at room temperature, i.e., apprroximately 20°C . III. RESULTS AND DISCUSSION All of the transition temperatures obtained from electrical resistance tests of the aged specimens and cold rolled specimens, including Ms, Mf, As, Af,Tp and Td temperatures, are summarized in Table 1 and Table 2 respectively. In this study several variables were changed in order to determine their effects on premartensitic as well as martensitic transformations. These variables include : aging temperature, aging time, cooling rate, degree of deformation ( i.e., thickness reduction percentage in cold rolling ) and thermal cycling. Finally, the effect of annealing following cold rolling was also examined. A) As-quenched specimen The As-quenched specimen was annealed at 900°C for 1 hr and then quenched into water without any further thermomechanical treatments. The electrical resistance vs. temperature curve of the As-quenched specimen for the first thermal cycle between +60°C and -196°C is plotted in Fig. 1(a). When this As-quenched specimen is cooled down from +60o C, the electrical resistance starts to rise at Tp (+30°C), while the first premartensitic transformation begins and ‘3 14 Table 1. Transition temperatures of annealed and aged specimens Specimen Ms Mf As Af Tp Td (°C) 900°C-1hr- -94 <-l96 -82 -52 +30 -70 water 900°C-1hr- -134 <-196 <-196 -196 +25 -67 air 800'C-1hr- -104 <-196 -77 -53 +22 -80 water 700°C-ihr- -96 <-l96 +36 -51 +24 -67 water 600°C-1hr- -87 <-196 -75 -43 +25 -64 water 600°C-4hrs- -66 -11o -56 -31 +20 +56 water 500'C-lhr- -44 <-196 +2 +15 +40 +18 water 500'C-4hrs- -13 -8 +17 +26 +33 +20 water soo‘c-lohrs- -14 -49 +22 +26 +30 +20 water 400°C-1hr- -38 <-196 <-196 -27 +49 +28 water 400°C-4hrs- -26 <-196 +3 +11 +52 +35 water 400°C-10hrs- -22 <-196 +8 +21 +54 +36 water 400°C-1hr- -4o <-l96 -26 -12 +54 +25 air 300°C-1hr- -125 <-196 <-l96 -125 +14 -34 Table 2. 15 Transition temperatures rolled and annealed" specimens of cold-rolled and "cold- Specimen Ms Mf As Af Tp Td (°C) C-R-10% -130 <-196 <-196 -160 --- --- C-R-20% -130 <-196 <-196 -155 --- -—- C-R-40% -164 <-196 <-196 -164 --- --- C-R-10%- -38 <-196 +2 +15 +38 +18 500°C-1hr- water C-R-20%- -42 <-196 -2 +16 +40 +19 500°C-1hr- water C-R-40%- -44 <-196 -18 +10 +41 +19 500°C-1hr- water 16 Electrical resistance ( Arbitrary units ) - r u r F uwoo o wee. Hawv. mHmonHHowH Hmmwmnmuom 32, <110>32 and <210>32, respectively. In addition to the exact Bragg diffraction positions from the 32 parent phase, diffuse streaks appear around 1/3 positions of the 32 reciprocal lattice. Since these streaks, arising from the incommensurate phase, are too faint and too diffuse to be indexed, they will be indexed by using different specimens. The second premartensitic transformation starts at Td ( -70° C ) which is determined as the inflection point of the resistance curve between Tp and Ms temperatures. Because Td is much lower than room temperature, the first premartensitic transformation observed under TEM is not obvious. This phenomenon is suspected of being responsible for the vagueness of the extra spots at 1/3 32 diffraction patterns. Below Td temperature, the commensurate phase, denoted by R phase, has been determined to have rhombohedral structure. The corresponding extra diffraction spots at 1/3 positions of the 32 reciprocal lattice from R phase can be seen in differently heat- treated specimens at room temperature. Martensitic transformation starts at Ms ( -94’ C ), where the 2(a) 2(b) Fig. 2(a) to 2(d). Transition electron micrograph and diffraction patterns of As-qienched specimen. 2(a) Bright field micrograph, mag.50,000x 2(b) [1111 32 zone diffraction pattern 20 2(c) 2(d) 2 (c) [1101 32 zone diffraction 2 (d) [2101 32 zone diffraction 21 resistance begins to decrease abruptly. Mf is reached when all the R phases transform into martensite. The Mf of As- quenched specimen is very colse to or slightly lower than liquid nitrogen temperature. The martensite structure has already been determined as distorted monoclinic ( 319') structure with the lattice parameter a = 28.89nm, b = 41.20nm, c = 46.22nm and p = 96.80”. Upon heating, austenite start temperature As ( -82°C ) is reached at the point of the abrupt increase in the resistance curve, and austenite finish temperature Af ( -52°C ) at the point of sharp decrease in resistance. Between cooling and heating curves, there exists a large electrical resistance hysterisis. After reaching Af temperature the specimen has returned to its parent phase. 3) Effect of aging temperature The effect of aging temperature on TiNi was studied by keeping other variables constant. The condition chosen was aging specimens at temperatures ranging from 300°C to 900°C for 1 hour, followed by water quenching. Figs. 3 through Fig. 8 are the electrical resistance vs. temperature curves for specimens aged at from 800° C to 300° C respectively. The transition temperatures from the firt cycle of the aged specimens are summarized in Table 1, including Ms, Mf, As, Af, Tp and Td etc, while they are compared in Fig. 9. For specimens aged from 800°C to 400°C, 22 the lower the aging temperatures, the higher the Ms, Tp and Td are. The shapes of the resistance curves of 800° C-lhr-water, 700° C-lhr—water and 600°C-1hr-water are very similar to that of the As—quenched specimen. The aging effect within this temperature range is the shifting of the resistance curves towards higher temperatures. The resistance curve of 500°C-1hr-water is different from those of specimens aged at higher temperatures. Precipitates are formed in the matrix with the size of 70° nm x 17nm. The formation of precipitates will lower the concentration of Ni in the matrix which could cause the increase of Ms temperature. The bright field and dark field micrographs of precipitates associated with its selected area diffraction micrograph are shown in Figs. 6(b), 6(c) and 6(d) respectively. Since Td is very close to room temperature ( Td = +18°C ), the matrix is in R phase which has rhombohedral structure. The composition of these precipitates is reported as about TiNi3, having hexagonal structure. There were three variants of precipitates with different orientations, and their longititudinal directions of precipitates have been determined ( See Fig. 6(d) ). The morphology of these cigar-like precipitates can be confirmed by the fringes at the edges of each grain. The resistance curve of 400° C-lhr-water is very special due to its heating curve. No obvious As can 23 Electrical resistance ( Arbitrary units ) - p b A“ _ Haonv IHoo mwo. u. mHmOdean Hmmwmdmsom 4m. dosomfldeHm ocfidm ow dvmfiamw OKOHm H voflsmmd +mq.n mum IHmmon. moonwemd moq.n|H:HI£man. 24 Electrical resistance ( Arbitrary units ) p p b n » P O O lHoo 0 .HA v wee. a. mHmOHHMowH Homwmnmuom 1 hr ) aging 38 Electrical resistance ( Arbitrary units ) . . s . . . on .6 .23 . IHoo mHo. Ho. mHmonHHowH Hmmwmnmsno