Titanium is an expensive yet highly desirable engineering material due to its excellent strength to weight ratio, corrosion resistance, and biocompatibility. The high cost of the material is due to inefficiencies in extractive metallurgy and difficulties caused by the inherent physical metallurgy of titanium. Emerging methods in extractive metallurgy promise significant reductions in the cost of production. Additionally, the native form of titanium produced from some of these processes is high purity powder. These factors combined have led to increased interest in titanium powder metallurgy techniques, challenging traditional methods of manufacture. This thesis broadly aims at developing novel technologies and techniques which pertain to cost-effective net-shape manufacturing of titanium components from powder precursors. Two long-standing problems in the field of titanium powder metallurgy have been approached.
Firstly, metal production via titanium tetrachloride thermochemical reduction produces magnesium or sodium chloride as a by-product. These by-products are difficult to remove from titanium. This, together with its volatile behaviour, causes significant defects and porosity during full density powder metallurgy. Rare earth oxides are known to react with chlorine producing rare earth oxychlorides. Trace additions of rare earth oxides, namely Y2O3, Nd2O3 and Er2O3, are shown here to lead to a dramatic increase in sintered density during liquid phase sintering. For example, compacts which swell, demonstrating excessive porosity due to the chloride impurities, sinter to almost full-density with 500 ppm Y2O3 additions. An equivalent improvement in sintered density may be attained by soaking compacts in a vacuum at an intermediate temperature for prolonged periods. The potency of rare earth oxides has also been demonstrated at long distances. The enhanced densification observed during sintering is due to the scavenging of chlorine by rare earth oxide additions. That is, rare earth oxides are involved in a chlorination reaction which reduces or eliminates entrapped gas otherwise preventing pore elimination. During solid state sintering, chlorine scavenging remains operative. However, since the influence of gas is small compared to the strength of the inter-particles bonds, the effect on densification is negligible in this sintering regime. The development of chlorine scavengers may lead to the development of impurity-tolerant PM alloys which utilise low cost, high chloride powders. The use of chlorine scavengers may also provide a novel solution to the limited weldability of titanium PM products. Broad applications may also exist if the chlorine scavenging behaviour is not limited to titanium.
The second part of the thesis focuses on full density net-shape processing of titanium powder – a universal yet challenging objective. Such a task requires excessive sintering or compaction conditions or additional stages of complex processing. The net-shape HIPping process, for example, necessitates ‘canning’ and ‘decanning’ which are slow and costly. Here we show that the need for a container during HIPping can be eliminated if the surface of the compact is densified in preference to the bulk. Such a phenomenon rarely occurs during sintering. However, under certain conditions sintered titanium compacts develop a densified surface several hundred micrometres thick. Sintering with a low volume fraction of liquid is a key condition for surface densification. Several other factors affecting surface densification were identified, including the need for compacts to contain some levels of chloride impurities. The mechanism of surface densification is linked to the depletion of volatile chloride impurities near the exterior of the compact. During heating, differences in the diffusion distance and slow mass transport through interconnected porosity lead to escape of gas occurring first near the edges of the compact. Regions near the surface become essentially chloride-free whereas pores in the bulk preserve entrapped chloride gas and cannot be eliminated during sintering. The slow mass transport of chlorides through interconnected porosity is also reported during vacuum purification of titanium sponge. Surface densification in titanium appears to occur by a similar mechanism to that reported elsewhere in copper and SiC-C compacts. Samples with a densified surface demonstrate a favourable response to containerless HIP. HIPped samples are virtually free of porosity. Furthermore, the densified surface has been shown to develop after very short sintering times, removing the need of a container during HIPping. The advantages of a combined process involving surface densification and containerless HIP are potentially useful for cost-effective net-shape manufacturing.