Strain-Induced Porosity Model

Advanced computer models improve the quality of titanium alloys.

August 1, 2006

AFRL scientists developed advanced computer models to improve the processing and quality of titanium alloys used in manufacturing gas turbine engine parts and critical structural components for military aircraft. AFRL transferred both the models and the basic materials knowledge to titanium mill suppliers to help them eliminate strain-induced porosity (SIP)—also known as cavitation—in billet products (see Figure 1) and finished parts. The models also increase product yield by reducing the amount of scrap material, which helps lower production costs.

Titanium is a very durable, low-density element (about 60% the density of iron) that gains considerable strength when processed as an alloy or mechanically altered via deformation processing. Because titanium alloys reduce weight and operate very effectively at low to moderately elevated temperatures, engineers have used them as replacements for iron alloys in aerospace applications for many years.

Corrosion-resistant parts and low-weight, high-specific- strength structures represent two typical applications for titanium and its alloys. In aircraft gas turbine engines, rotating components such as turbine disks and blades require titanium alloys in order to maximize strength-to-weight ratio, metallurgical integrity, and reliability at service temperatures. These alloys must exhibit good fatigue resistance and low creep rates. Stringent user requirements ensure controlled homogeneous microstructures and freedom from imperfections. Melting-related imperfections include alloy segregation, high- or low-density inclusions, and ingot porosity, while flaws associated with thermomechanical processing include SIP, shear bands, and undesirable residual stresses.

SIP is the undesirable formation of cavities that form during hot working of metallic materials (see Figure 2). The cavities usually nucleate at grain boundaries (the boundaries between crystals) as a result of local stresses and strains induced by sliding that occurs along the boundaries at hot-working temperatures. Depending on the precise state of stress and strain, the cavities grow to various sizes (from nano- to micrometers) and, in extreme cases, may lead to fracture. Although internal, micrometer-scale cavities are extremely difficult to detect using nondestructive techniques, avoiding cavity formation and growth during hot working is nevertheless important for avoiding the harmful effect of SIP on subsequent processing or service.

Researchers focused on three different approaches for modeling cavitation during hot working of ductile metals: (1) phenomenological, (2) mesoscale mechanistic, and (3) microscale mechanistic. Phenomenological approaches relate the occurrence of damage and gross fracture in complex stress states to measurements made under a simple stress state, such as uniaxial tension. In contrast, mesoscale mechanistic models examine the plastic growth of individual cavities and their coalescence. Researchers can obtain reasonable estimates of hot ductility from mesoscale modeling, but these techniques do require some assumptions regarding cavity nucleation. Fundamental microscale mechanistic models describe the mechanisms of early-stage cavity growth to provide a basis for quantifying nucleation-type behavior.

AFRL scientists began working on cavitation models in 1997; this initial research resulted in the successful development of the phenomenological and mesoscale mechanistic models. AFRL transferred these models to titanium mill suppliers, who are using them to modify and improve their production practices. Quality improvements in titanium and its alloys will also benefit commercial aircraft and other key industries.

Dr. Lee Semiatin and Dr. Peter S. Meltzer (Anteon Corporation), of the Air Force Research Laboratory's Materials and Manufacturing Directorate, wrote this article. For more information, contact TECH CONNECT at (800) 203-6451 or place a request at http://www.afrl.af.mil/techconn_index.asp . Reference document ML-H-05-06.