A novel combinatorial approach to the development of beta titanium alloys for orthopaedic implants

In recent years there has been a significant thrust directed towards the development of novel implant alloys based on β-Ti. Two recently developed and promising biocompatible β-Ti alloys are Ti-35Nb-7Zr-5Ta and Ti-29Nb-4.6Zr-13Ta. While both these alloy compositions, based on the quaternary Ti-Nb-Zr-Ta system, are promising, there is still a tremendous scope for improvement in terms of alloy design in this and other systems via optimization of alloy composition and thermo-mechanical treatments. Here a novel combinatorial approach has been used for the development of implant alloys with optimized compositions and microstructures. Using directed laser deposition, compositionally graded alloy samples based on the Ti-Nb-Zr-Ta system have been fabricated. These samples have been heat-treated to affect different microstructures in terms of the volume fraction and distribution of the α phase in the β matrix as a function of composition. Subsequently, composition-specific indentation-based hardness and modulus information has been obtained from these samples to construct a database relating the composition and microstructure to the mechanical properties. These databases have been used to train and test fuzzy-logic based neural-network models for predicting the mechanical properties. The trained models have also been used to predict the influence of different alloying additions on the hardness and modulus. These predictions have subsequently been verified by detailed experimental characterization, shedding light on the factors influencing the strength and modulus in these alloys. Such modeling approaches for the development of novel implant alloys can be highly beneficial since they offer the possibility of identifying promising compositions without the necessity for extensive experimental test cycles.