Price, Timothy Simon
Cold gas dynamic spraying of titanium coatings.
PhD thesis, University of Nottingham.
Cold gas dynamic spraying, CGDS, is a relatively new technique used to deposit materials onto the surface of a substrate. It differs to the majority of other thermal spray techniques as the substrate and particles are not exposed to high temperatures during the spraying process. This makes CGDS particularly advantageous for spraying materials such as titanium which react at high temperature. The aim of this project was to investigate the potential use of titanium coatings by CGDS as a surface treatment for medical prostheses. Titanium powder was deposited onto Ti6A14V substrate using helium gas at room temperature. Titanium coatings were also produced by a competing spray technique, shrouded arc spraying, to allow for a comparison of the two techniques to be made. Mechanical properties of the titanium coatings were measured and the influence of subsequent heat treating on the mechanical properties was also investigated. The coatings were characterised by investigating properties such as their bond strength, residual stress and stiffness, as well as the influence of the cold spray titanium coating on the fatigue life of a sprayed substrate. Scanning electron microscopy and optical microscopy techniques were also used to characterise the deposits. To further develop and optimise the CGDS process aluminium and copper coatings were also deposited and their mechanical properties compared to that of the titanium deposits. Additionally particle image velocimetry, PIV, was used to improve general understanding of the cold spray process and the effect that spray parameters have on the particle impact velocity.
In the case of fatigue endurance limit and despite a compressive residual stress state measured in the titanium CGDS coatings, a 15% reduction in fatigue endurance limit was observed following the application of a CGDS titanium coating to the as received substrate, but no significant reduction was observed on its application to the grit blasted substrate. By four point bend testing it was observed that the ratio of the modulus of the titanium deposit to that of the corresponding bulk material (the modulus ratio) was 0.17, significantly below unity. For copper and aluminium, also deposited by cold spray, a modulus ratio of 0.41 and 0.16 was observed. The volume fraction and aspect ratio of porosity in each deposit was measured by SEM. However, an Eshelby equivalent homogeneous inclusion model supplied with these data was not able to predict the low modulus ratios observed. Instead, imperfect inter-particle bonding within the deposit (akin to through-thickness cracks) is the source of the low modulus ratios, with the tenacious oxide on the titanium and aluminium powder particles being more effective at preventing oxide disruption and formation of metallurgical bonds between particles upon impact, hence the lower modulus ratio for these material types.
A method was developed to visually show the imperfect inter-particle bonding expected to be found within the CGDS deposits. Although porosity levels are low within cold spray deposits, individual particles are found to not be well bonded to each other which results in low coating moduli. By increasing the primary gas pressure in the cold spray process an increase in the degree of inter-particle bond formation occurred.
Heat treating of the titanium coating at 1150 °C was found to improve the bond strength of the coating but drastically reduced the fatigue endurance limit. The same heat treatment temperature was found to increase the modulus ratio of the coating to 0.36 and this is attributed to a greater level of inter-particle bonding within the titanium coating occurring by diffusion bonding. However a reduction in the fatigue endurance limit was observed. This is most likely due to phase changes of the Ti6A14V substrate, the stress relaxation nature of heat treating and oxygen embrittlement occurring, despite an argon furnace used.
Diagnostic tools such as particle velocity measurements are also used to gain a further general understanding of the CGDS process. Particle velocity measurements were made for titanium and copper powders using helium and nitrogen gas and a range of spray parameters. Particles of less than 9 pm in diameter were found to have the slowest particle velocities for both the copper and titanium powders. For the titanium powder up to a 200 m s-1 variation was found between the smallest and largest sized particles. This knowledge may be used in the future to optimise the size distribution of a powder feedstock prior to being used for cold spray deposition. Overall the titanium powder produced the highest particle velocities compared to the copper powder due to its lower density and therefore being easier to accelerate by the gas flow. A particle model was used to predict particle velocities. Generally the model predicted higher particle velocities then the maximum measured particle velocities and this is due to the model not taking into consideration of particles interacting with one another, the external walls of the nozzle, or the external atmosphere. These factors would all lead to particle deceleration.
In comparison to titanium coatings produced by the shrouded arc process, the titanium CGDS coatings performed admirably. The shrouded arc coatings had a considerably lower fatigue endurance limit, due to tensile residual stresses within the coating, forming as particles cool on impact. However, the shrouded arc coatings showed a higher modulus ratio of approximately 1/3, which is comparable to coatings produced by other thermal spray methods.
Thesis (University of Nottingham only)
||Metal spraying, coating processes, gas dynamics
||T Technology > TP Chemical technology
||UK Campuses > Faculty of Engineering > Department of Mechanical, Materials and Manufacturing Engineering
||13 Mar 2012 11:34
||13 Sep 2016 11:27
Actions (Archive Staff Only)