Optimizing strength and fracture toughness of a cast titanium alloy through heat treatment and microstructure control

by Robinson, Amy C.

The relationship between the microstructure and tensile ductility and fracture toughness for cast Ti-5111 was determined and compared to that of hot-rolled and annealed Ti-5111. Graphite mold cast Ti-5111 plate material was examined in the as-received condition and after six different heat treatments involving elevated temperature anneals followed by an air or furnace cool. Three investment cast Ti- 5111 plates were also examined after annealing followed by either a fan cool, air cool, or furnace cool. All castings developed a lamellar colony microstructure consisting of aligned lamellae of alpha and beta phases. Altering the cooling rate from the annealing temperature had the most influence on the microstructure such that plates with a slower cooling rate typically developed coarser grain boundary alpha, larger alpha colonies, thicker alpha laths, and greater volume fractions of alpha phase. The average prior beta grain size for the graphite mold cast specimens ranged from 920 µm to 1360 µm, while that for the investment cast specimens was approximately 1750 µm. The tensile behavior of the castings was characterized by a crack initiation and propagation process where the ductility was often limited by the strain required to initiate a large crack. The cracks formed along planar slip bands that crossed alpha colonies or in some cases, entire prior beta grains. Thus, reducing the alpha colony size and prior beta grain size should improve the casting ductility by limiting the length of slip-induced cracks. Due to the large grain and colony sizes present in the castings, the strength and ductility was observed to be sensitive to specimen size such that a smaller tensile diameter (i.e. 3.2 mm as compared to 12.5 mm) iv decreased the tensile and yield strengths due to the high fraction of large grains located on the specimen surface that can yield by predominantly single slip. The scatter in ductility values in the smaller specimens was significantly greater as a result of fracture controlled by a crack initiation and propagation process within a single grain that comprises a large fraction of the specimen cross-section. Thus, once a large crack initiated, minimal additional strain was required to propagate the crack. Both intrinsic and extrinsic toughening mechanisms were apparent in the fracture toughness study of the castings. The fracture initiation toughness was enhanced by secondary cracking and significant blunting at the crack tip as evidenced by the presence of strain-induced void formation within a large process zone. Large alpha colonies located at the transition between the fatigue pre-crack and tensile crack growth regions limited the fracture initiation toughness by promoting easy crack growth along a significant fraction of the crack front. Thus, limiting the alpha colony size should enhance the fracture initiation toughness. The best crack propagation resistance (tearing modulus) was observed from specimens with large alpha colonies and large prior beta grains. Enhancing the size of these features increased the surface roughness, and consequently the tearing modulus, due to greater crack deflection, crack bifurcation, and shear ligament toughening from the larger alpha colonies and prior beta grains. Crack bridging by the ductile beta phase was also observed and should enhance the tearing modulus. When compared to the hot-rolled and annealed plate, the graphite mold castings exhibited better fracture initiation toughness and crack propagation resistance. v However, the wrought plate maintained relatively good fracture initiation toughness and crack propagation resistance as a result of the continuous ductile beta matrix present in the microstructure. vi