A new method to capture and evaluate the condition of thermal sprayed particles has been developed by using an agar gel target. HVOF sprayed Hastelloy C particles were collected by a gel target placed at the substrate position, i.e., 380mm downstream from the spray nozzle exit. In the surface layer of the target, a large number of fine particles were observed, whereas in the deeper part, globular particles were trapped. The ratio of particles in the surface layer with respect to the deeper part changed by spraying parameters. Furthermore, particles in the target were separated by cutting the gel, collected after resolving the agar, and then observed by SEM. Particles collected from the target's surface layer were fine particles (under 10µm) and fragments of dendritic crystals. These collected from the deeper part were mostly unmelted particles, some of which exposed dendrites. Results obtained by other techniques such as splat observation and in-flight diagnostics are compared with these results. It was concluded that by capturing thermal sprayed particles with a gel target, it is possible to visualize and quantify the melting condition of HVOF sprayed particles.

In plasma spraying, the dynamics of droplet deformation and cooling on the substrate determines the geometry of the resulting splat, the porosity formation and the quality of contact between the splat and the underlying layer, thereby controlling the properties of coatings. The process of splat formation depends on the droplet velocity, size, molten state, chemistry and angle of impact, and on the substrate material, roughness, temperature and surface chemistry. This paper concentrates on investigating the way the droplets impact onto the substrate. A fast-imaging camera is used to observe the impact of plasma-sprayed alumina droplets on cold and hot substrates. Three modes of impact are observed: splashing, deposition or rebounding. A droplet is said to splash when it disintegrates totally or partially into secondary droplets after impacting onto the surface while it is said to deposit if the liquid material spreads and solidifies on the surface with a minimum splashing. The results are expressed in dimensionless form as a correlation between the impact mode and the Sommerfeld parameter K. The latter quantifies the importance of the particle parameters at impact (temperature, velocity and size) and viscous and surface tension forces. It is found that, when K is higher than 57.7, the splashing mode prevails as already found in a variety of different fields for droplets impacting on surfaces.

The effect of substrate characteristics on the formation of plasma-sprayed alumina splats was studied using both experiments and numerical simulation. Knowledge of the particle and substrate conditions is critical in understanding coating formation and in validating computational models. The size, velocity and temperature of the alumina particles prior to impact were measured using a particle in-flight diagnostic system. Experiments were performed on two substrate materials: stainless steel and glass. Substrate temperatures were varied in a range of 20-500°C and controlled with an electric heater. For each substrate material, a transition temperature was observed above which there was no fingering/splashing and the splats had a circular disk shape. A 3D computational model of free surface flows with heat transfer and solidification was used to simulate the impact of alumina particles in conditions given by the experiments. The splat shapes from numerical model were comparable to those of the experiments for hot stainless steel substrate. For a cold substrate, the numerical model did not show any fingering/splashing. In the experiments, however, we observed two types of splat shapes: intensive splashing with no central core and circular disk splat. Substrate surface contamination, not considered in the numerical model, was the probable cause of droplet splashing on the cold substrate.

Investigations have been carried out to study the influence of Atmospheric Plasma Spray process (APS) on the in-flight oxidation of pure iron particles. After collecting the molten droplets in-flight, XRD, SEM and Mössbauer spectroscopy are used to determine the amount and distribution of formed oxides. The results indicate that the Wüstite, Fe 0.95 O, is the only oxide formed during the APS. In the jet core and at the beginning of its plume, oxidation is controlled by convection within molten droplets and then for the downstream of the plasma by diffusion where the solubility of oxygen through the external oxide layer governs the growth of Wüstite. Calculations have shown that the convective movement is due to the drastic velocity difference between the plasma jet and particles. Wüstite granules can be distinguished within the particles due to the immiscibility between Fe and FeO in liquid phase. This oxide phase represents about 13 wt% of the collected particles at 100 mm stand-off distance in an Ar-H 2 plasma jet (50:10 SLM) with 18 kW effective power. The amount of oxide decreases when the H 2 volume percentage of the plasma, the internal diameter of the anode nozzle and the effective power increase. Sessile drop studies of molten iron on ceramic substrates are carried out to simulate the wetting of oxidized iron particles during coating formation. It is found that the oxidation state of iron particles during APS has a significant effect on the observed contact angles. A strong decrease of the contact angle is observed in the case of oxidized iron particles. In-flight oxidation of iron particles allows a better splat covering.

Fundamental aspects of a plasma sprayed cast iron coating on an aluminum alloy substrate are investigated in the present study: focusing on the effects of preheat substrate temperature (T S ) and chamber pressure (P C ) on the splat morphology, the adhesive strength of splats, the formation of a reaction layer and graphite. Splash-type splats appear at low T S but disk and star-shaped splats arise at high T S . Deformed substrate ridges, mainly due to the slight surface melting, are formed adjacent to the splat periphery at high T S . At low T S , pores are observed at the splat/substrate interface, which cause a decrease in the adhesion of splats. In contrast, a reaction layer composed of iron, aluminum and oxygen is ready to form at high T S . The amount of graphitized carbon increases in cast iron splats with T S . At a low P C of 26.3 kPa, disk-type splats are in the majority at a constant T S of 473 K. As P C increases, star-shaped splats appear along with disk splats. The flattening ratio of disk splats decreases with the increase of P C , because of a decrease in the kinetic energy and temperature of molten droplets. An interfacial oxide layer composed of iron, aluminum and oxygen is ready to form at high P C . The number of pores intensively increases with P C , which leads to a decrease in the adhesive strength of splats. The amount of formed graphite in cast iron splats slightly increases with P C , however, that of a rapidly solidified phase of Fe-Si-C decreases because of lowering of the solidification rate.

New method for characterization of coating microstructures and for evaluation of coating property by means of surface morphology has been proposed. In this paper, the distribution of shape and dimensions of splat was examined using quantitative analysis of scanning electron microscope images from the surface of spray pattern as well as the surface of coating. Results obtained in this study indicate that it is necessary to analyze the spray pattern as well as the surface morphology in order to estimate the coating property by means of the distribution of splat which composes the coating. Moreover, the splats, which are in the interface between the substrate and the coating, should have the same morphology as those of the coating surface. Therefore, the analysis of the surface morphology is important even for the evaluation of coating adhesion behavior.

In the collision of a liquid droplet onto a flat surface, a splashing parameter (K=We 0.5 Re 0.25 ) has been used to evaluate the flattening behavior, and a critical value of K (K C ) was introduced as a criterion for splashing. In order to evaluate K of thermal sprayed particle, in-flight measurement for the velocity and temperature of the particle was conducted in this study. As a flattening pattern of the thermal sprayed particle changed significantly with a substrate temperature, the transition temperature (T t ) was also measured by changing the substrate temperature. Both K and T t showed a tendency of monotonous decreasing with increasing the spray distance and a strong linearity was recognized in a K-T t relationship. This straight line corresponds to a critical value for the splashing. However, as the influence of substrate temperature on the flattening is essentially independent of K, a new criterion for the splashing in the flattening of the particle was proposed in the study. That is, a splashing parameter on flattening was proposed, which considers the ratio of the flattening velocity to the impact velocity of the particle.

It is well accepted that the morphology and microstructure of the splats have a strong influence on the characteristics and properties of thermally sprayed coatings. McPherson has made pioneering and outstanding contributions in the above area, especially for plasma sprayed coatings. Recently, splat morphology - microstructure - properties correlation has also been attempted in the case of HVOF thermal spray coatings. However, only limited data is available in the case of detonation sprayed coatings inspite of the fact that DS coatings have been available commercially for a long time. In the present work, the influence of particle velocity and temperature on the splat morphology and also area coverage of the splat has been studied for detonation sprayed Al 2 O 3 particles on a mild steel substrate. Further, the effect of two detonation spray process variables namely, oxy fuel ratio and shot frequency on splat morphology and splat area coverage has been evaluated. The above correlation has then been utilized to understand the variation of deposition efficiency of detonation sprayed Al 2 O 3 coatings on mild steel as a function of spray process parameters.

In the present work ceramic particles (Al 2 O 3 , YSZ) are sprayed onto steel substrates using a radio frequency inductively coupled atmospheric plasma spray process (IC APS). Because of the low plasma velocity and the large plasma volume large particles can be completely melted. The particles reach the substrate with low velocities (in the order of 10 m/s). So, a special kind of deformation can be observed. Some characteristic values of impact and deformation are also quite different from some other thermal spraying technologies. Of course, that has an strong influence on the coating properties. It is shown, that a high kinetic energy of impinging particles is not an essential assumption for a high bond strength and a low porosity of the coatings. IC Plasma sprayed particle splats are investigated and compared with DC and HVOF sprayed ones. The influence of the particle impact and deformation on the coating properties is demonstrated. It is shown, that in spite of the low particle velocities coatings can be sprayed by IC APS with comparable quality, but with quite different coating properties such as the crystalline structure.

As underlined in 1981 by Mc Pherson, thermo-mechanical properties of plasma-sprayed coatings depend not only on the way particles flatten and resulting splats solidify and cool down, but also on the thermal history of particle layering at the same location. To illustrate what is our present knowledge in that field, plasma-sprayed alumina coatings will be considered through modelings and measurements. The first part of this paper discusses the phenomena linked with particle impact and splat formation: splashing, spreading, solidification and grain growth, angle of impact in conjunction with particle parameters at impact and substrate surface parameters (chemistry, phase structure and roughness, temperature). The second part examines splats layering. It addresses the influence of plasma jet heat flux, relative velocity torch-substrate, powder flow rate and deposition efficiency on splat time-temperature evolution and resulting quenching stress, coating adhesion/cohesion and microstructure. The shadow effect when spraying off normal angle is also discussed. The last part deals with the effect of the successive cooling and reheating of passes on coating properties, and condensation of the vapor issued from evaporating particles.