INVESTIGATION OF DUAL STAGE HIGH VELOCITY OXYGEN FUEL THERMAL SPRAY SYSTEM

Transcription

1 International Conference on Applied Energy ICAE 2013, Jul 1 4, 2013, Pretoria, South Africa Paper ID: ICAE INVESTIGATION OF DUAL STAGE HIGH VELOCITY OXYGEN FUEL THERMAL SPRAY SYSTEM Mohammed N. Khan, Tariq Shamim Mechanical Engineering, Masdar Institute of Science and Technology. P. O. Box 54224, Abu Dhabi, UAE. ABSTRACT The high velocity oxygen fuel (HVOF) thermal spray coatings are used to protect the surfaces from deterioration. The base material surface properties can be modified to achieve the longevity of the product. Besides spraying material, the coating quality depends greatly on the gas and particle dynamics. The coating quality is also affected by the particle temperature, particularly for the materials which are sensitive to temperature such as titanium and copper. As the temperature of the gas phase is high, the material gets melted and oxidized before it reaches the substrate. To avoid this problem, a dual stage thermal spray system which is also called as warm spray has been developed for coating temperature sensitive materials. This process involves making coatings by high velocity impact of powder particles heated to temperatures below their melting point. The advantages of a warm spray process include an easy control of particle oxidation and production of various coating structures by controlling the particle velocity and temperature. The particle temperature can be controlled by varying the coolant flow rate in the mixing chamber. The present study investigates the effect of various geometric parameters of a dual stage thermal spray gun by developing a comprehensive mathematical model. The objective is to develop a predictive understanding of various design parameters. Conservation of mass, momentum and energy of reacting gases were taken into account while developing the model. Due to low particle loading, the particle phase was decoupled from the gas phase. The results demonstrate the advantage of dual stage system over single stage system especially for the deposition of temperature sensitive materials. Keywords: thermal spray, high velocity oxy fuel, coating, simulation NONMENCLATURE A empirical coefficient in eddy dissipation model A p particle cross sectional area B empirical coefficient in eddy dissipation model c turbulent model constant C D drag coefficient C p specific heat at constant pressure d p particle diameter external force h c convective heat transfer coefficient H overall enthalpy k turbulent kinetic energy m p mass of particle Nu Nusselt number p gas pressure P k production rate of turbulent kinetic energy P r Prandtl number R gas constant R F volumetric fuel consumption rate R e Reynolds number S source term t time T temperature u i velocity in the i direction instantaneous gas velocity vector x i spatial coordinate in the i direction Y fluctuating dilation in compressible turbulence to the overall dissipation rate. Greek variables α Inverse effective Prandtl number ε rate of turbulent kinetic energy dissipation λ thermal conductivity µ molecular viscosity µ eff effective viscosity

2 Γ k Γ ε diffusion coefficient of k diffusion coefficient of ε ρ density σ turbulent model constant Subscripts F fuel g gas O oxidant p particle P combustion product t turbulent 1. INTRODUCTION In many industrial applications, components are subjected to heat and corrosion and may require surface finish and protection from the environment. To achieve these, coatings of different materials are applied on the components using different coating methods. High velocity oxygen fuel thermal spray process is one of these techniques which is used in various engineering applications, particularly in aerospace and automotive industries. Titanium and titanium alloys are light weight, possess very high strength, highly corrosion resistant and are known to be the excellent candidates for aerospace applications [1]. Biocompatibility and osseointegratability of titanium within human body make it highly useful in biomedical implants [2]. In spite of being costly, the above mentioned properties and applications make the titanium a valuable material within engineering and surface coating industries [1, 3 6]. Titanium being a temperature sensitive material has a strong affinity towards oxygen [7]. Therefore, titanium cannot be used with a conventional high velocity oxygen fuel thermal spray since the temperature generated during the process is above 3000 K. This high temperature will melt the particles completely before they reach the substrate resulting in a low quality coating. These melted particles undergo oxidation during the flight and the oxidation increases exponentially when the particle is heated beyond 900K [8]. Hence, control of particle temperature and the oxidation becomes important. This could be achieved by using cold gas dynamics spray (CGDS) [9] but this method results in low deposition efficiency and high porosities within the coating due to the low temperature associated with this process, which hinders the plastic deformation of the particles. The void in the temperatures from both high velocity oxygen fuel system and cold spray system was overcome by a recently developed technique called dual stage thermal spray system which is also called as warm spray system. This was first patented by Browning [10] and later developed as warm spray by many researchers [11 14]. The schematic of a dual stage spray gun is shown in Figure 1. The gas phase temperature in the warm spray can be controlled in the range of around K depending on the feed material while keeping the particle velocity similar to that of HVOF process. The Dual stage gun has a combustion chamber followed by a convergent nozzle. A mixing chamber is introduced between convergent nozzle and the converging diverging nozzle followed by a barrel. The temperature is controlled by varying the nitrogen mass flow rate in the mixing chamber. The combustion gases produced mix with nitrogen in the mixing chamber and accelerate to supersonic speed through a convergingdiverging nozzle. The gas leaves the barrel with a supersonic speed. The metal powder particles are injected at the upstream of the barrel. The particles get heated by the gas and are accelerated to high speeds in the downstream direction. After exiting the barrel, the particles hit the substrate with a high impact velocity and adhere to it forming a coating. The thermal spray systems involve complex processes such as turbulent flow, heat transfer, multiphase flow (gas Figure 1 Schematic of Dual Stage Thermal Spray Gun 2 Copyright 2013 by ICAE2013

3 and solid particles), chemical reactions and supersonic/subsonic flow transitions [15]. The thermal spray processes can be modeled in two parts: modeling of the gas phase and modeling of particle phase. The particle and the gas field can be decoupled because of negligible effect of particle phase on the gas phase due to low particle loading. Hence, both the parts can be modeled separately. Many computational fluid dynamic simulations have been performed for both gas fueled and liquid fueled single stage spray systems [16 18]. On the other hand, very few computational studies have been performed on liquid fueled warm spray systems only but with focus on controlling the particle temperature by controlling the coolant mass flow rate in the mixing chamber. Tabbara et al., [2] have modeled the titanium particles in warm spray by considering gas particle interaction and phase change within the particles. They also showed how the nitrogen flow rate in the mixing chamber affects the titanium particle temperature and velocity. The present study is performed by employing a central cooling chamber concept developed by [12, 18, 19] and the modeling process developed by Dolatabadi et al., [20]. The physical properties and microstructure of the coatings depend greatly on the particle condition at the substrate and the particle size. These, in turn, are governed by the gas dynamics and parameters such as the oxygen/fuel ratio, gas jet, particle injection method, gun design, etc. [21]. Despite the gun design being a very important factor in determining the gas and particle phase, it has not got much attention. The present study investigates the effect of geometrical parameters such as length and diameter of combustion chamber, mixing chamber and the diverging section of the convergingdiverging nozzle on the gas phase and particle phase. 2. MODEL DEVELOPMENT AND MATHEMATICAL FORMULATION Due to low powder feed rate, the mathematical model is developed by decoupling the gas phase from the particle phase. This allows modeling of the gas phase and the particle phase separately. The model is solved by taking into account the conservation of mass, momentum, energy, turbulence and equilibrium chemistry. The solution obtained for the gas phase will be used to model the particle to determine its velocity, temperature, position and size distribution because the particle phase will not affect the gas phase. For solving the high speed, compressible and Newtonian flow gas phase, ideal gas assumptions and Eulerian formulation are assumed, whereas Lagrangian formulation is adopted for the particle phase. 2.1 Gas phase dynamics The governing equations for the gas phase are the contin uity, momentum and energy. The pressure and density are linked by ideal gas equation of state. The flow in this process is viscous, compressible and turbulent. The turbulence is modeled by using RNG k ɛ model with nonequilibrium wall function treatment which enhances the wall shear prediction and heat transfer. The governing equations are listed as follows: Continuity: (1) 0 Momentum: t ρu (2) ρu x u p S x μ x u u 2 x x 3 μ u δ x Energy: ρh ρu t x H p t λ T μ h (3) x x Pr x S u x μ u u 2 x x 3 μ u δ x μ k x Equation of state: pρrt (4) where μ μμ (effective viscosity) (5) μ ρc (Eddy viscosity) (6) Hh u u k (Overall enthalpy) (7) RNG K ɛ turbulence model: Turbulent kinetic energy: ρk ρu t x k α x μμ k P x ρεy Rate of turbulent kinetic energy dissipation: (9) ρε ρu t x ε α x μμ ε ε x k c P ρc ε R The values used for the constants are c 1 =1.42, c 2 =1.68, c µ = P k is the production rate of turbulent kinetic energy and is given as: P μ u u u 2 x x x 3 ρk μ u u (10) x x Combustion The combustion of hydrocarbons is a complex phenomenon. High temperatures are generated which will (8) 3 Copyright 2013 by ICAE2013

4 dissociate CO 2 and H 2 O into a number of low molecular weight species due to strong thermal atomic vibration [22]. Therefore, the dissociation of the major species should be considered otherwise the combustion model will overpredict the temperature [16]. In this paper, eddy dissipation model is used which assumes the reactions to be infinitely fast and the reaction rate is limited by the turbulent mixing rate of fuel and oxidizer. Therefore, Arrhenius Kinetic rate data is not needed since the model describes the limiting rate which is validated by [20]. The volumetric consumption rate of fuel is given by R ρε (11) k A min m, m,b m s s where S n M /n M (12) S n M /n M (13) Here, A and B are empirical coefficients with default values of 4 and 0.5 respectively. The chemistry is taken from [23] which was derived by first calculating the chamber pressure by using 1 D process [24] and then an equilibrium stoichiometry is obtained by using a chemical equilibrium [25]. The chemistry model involves nine gas species: C H,O,CO,CO,H,H,H O, O and OH and the chemical equation is as follows. C H 4.307O 1.903CO 1.097CO 0.382H 0.432H 2.004H O 0.388O 0.745OH 0.692O (14) 2.2 Particulate phase dynamics Due to low particle loading, the particle phase is decoupled from the gas phase. Hence, there will not be any influence of particles on the gas but there will be influence of gas on the particles through momentum transfer and heat transfer. The motion of the particles is modeled by the Lagrangian particle tracking method. The drag force on the particle is dominant [21] and as a result the particle motion can be described by the following equation dv m dt 1 2 ρ (15) A C U UU UF where is the particle mass, and are instantaneous particle and gas velocities, is gas density, is the particle cross section area, is the drag coefficient and represents the external forces such as gravitational forces. The drag coefficient depends on the particle Reynolds number, particle Mach number, particle surface roughness and the rotation of particles. The most important parameter is local Reynolds number [20] Re. (16),Re C Re , Re 10 where, the Reynolds number is given as U U (17) Re ρ d μ Here, is the viscosity of the gas and is the particle diameter. The particle Biot number during a spray process is usually less than 0.1 [22]. Accordingly, the temperature gradients inside the particle and the internal resistance during heating can be neglected. Therefore, the particle is assumed as a lumped system. The energy equation for a single particle neglecting the heat transfer through radiation, is given by dt m C dt A (18) h T T where, is the specific heat of the particle and is the convective heat transfer coefficient defined as h Nu (19) d where, is the Nusselt number, which is given by the following correlation [26]. Nu Pr. Re. (20) 2.4 Computational domain and boundary conditions A schematic diagram of the dual stage thermal spray gun based on the study conducted by Katanoda et al [18] along with the dimensions is shown in the Figure 2. The figure highlights the fuel oxygen inlet (a), combustion chamber (b), convergent nozzle (c), nitrogen inlet (d), mixing chamber (e), converging diverging nozzle (f), carrier gas and particles inlet (g), barrel (h) and ambient (i). Since the problem is axisymmetric, only half of the two dimensional domain is modeled. The computational domain for the base case is divided into number of regions and the selected mesh is shown in Figure 3. The combustion chamber consists of 120x20 grids, converging nozzle consists of 36x20 grids, mixing chamber has 46x50 grids, converging diverging nozzle consists of 118x30 grids and barrel consists of 240x30 grids. In the outside domain, the horizontal mesh includes 150 uniform grids for the first 62mm followed by 150 stretched grids. In the vertical mesh, there are 30 uniform grids for the first 5mm followed by 50 stretched grids. The grid independence was performed by creating nine meshes with different number of elements. The elements were varied by varying the number of intervals on every edge by a specific percentage. The current mesh size is selected such that the pressure, velocity and temperature are independent of the mesh size. The boundary conditions are all kept similar to those in reference [20]. There are three inlets and all the gas inlets are assumed to be at a temperature of 300K. The walls are smooth and are assumed to be at a temperature of 350K. The pressure at the open boundaries is assumed to be 1 atm. The titanium particles are used in this study with 4 Copyright 2013 by ICAE2013

5 particle diameter d p = 15µm, C p = J/kg.K, ρ p = 4850 kg/m 3 and k p = 7.44 W/m.K. 3. MODEL VALIDATION For validation, the current model is applied on the single stage domain with MCrAlY particles as described by Dolatabadi et al., [20]. The results obtained for particle temperature and velocity are compared with the numerical and experimental results of [20] and are shown in Figure 4. The figures shows that the current numerical results are in better agreement with the experimental results as compared to the numerical results of Ref. [20]. After validating the model, the same domain is used but this time titanium particles are introduced instead of MCrAlY particles used by [20] to investigate the use of a single stage spray system for a temperature sensitive material such as titanium. The titanium particle temperature is plotted in Figure 5, which shows that by using single stage system, the titanium particle temperature becomes very high, almost reaching its melting point of 1943K. Recall that for a good quality coating, the particles should reach their plastic state rather than the melting since melting increases the rate of particle oxidation which is not desired. The same model with titanium particles is then employed to simulate a dual stage thermal spray gun and the particle is also shown in Figure 5. In the dual stage system, the maximum particle temperature achieved is about 1198K which brings the particle reach its plastic state. Figure 2 Schematic of a Dual Stage Thermal Spray gun depicting fuel oxygen inlet (a), combustion chamber (b), convergent nozzle (c), nitrogen inlet (d), mixing chamber (e), converging diverging nozzle (f), carrier gas and particles inlet (g), barrel (h) and ambient (i) a) b) c) Figure 3 Computational mesh for a) combustion chamber and converging nozzle, b) mixing chamber and convergingdiverging nozzle and c) barrel and open boundaries 5 Copyright 2013 by ICAE2013

6 a) b) Figure 4 Validation of current results with experimental and numerical results of [20] a) Particle temperature b) Particle velocity Figure 6 Variation of gas phase and particle phase temperatures and velocities for the base case single stage system which is around 600 m/s demonstrating the advantage of using dual stage system. Furthermore, particle phase have an increasing trend when compared to decreasing trend of the gas phase after the barrel exit. This is due to the fact that the particles have lower temperature than gas phase and hence, continues to gain temperature even when the gas phase is losing temperature. The effects of the important geometrical parameters such as lengths and diameters of combustion chamber, mixing chamber and diverging section are described in the subsequent sections. Figure 5 Variation of titanium particle temperature in single and dual stage thermal spray systems 4. RESULTS AND DISCUSSIONS The model simulation results for the dual stage thermal spray system are shown in Figure 6. The gas phase temperature and velocity are plotted from the barrel entrance so that the particle temperature and velocity can be studied relative to the gas phase. The figure shows that shocks occur at two places, one at the entrance of the barrel and another one at the exit of the barrel. The peak particle temperature achieved is at the outside of the barrel which is around 1198K as opposed to 1950K in single stage system. The particle velocity is almost similar to that of the 4.1. Effect of combustion chamber diameter The combustion chamber diameter does not have significant effect on the gas phase and particle phase as shown in Figures 7 8. The combustion chamber diameter was varied from 20mm to 50mm. The results show that there is a slight decrease in the gas phase temperature as the combustion chamber diameter increases. This is due to the small heat loss that occurs due to increase in the combustion chamber area. As the area is increased, more heat transfer occurs resulting in relatively lower temperature. The gas phase temperature affects the particle temperature. Slight decrease in the particle temperature has been observed as the combustion chamber diameter increases. The maximum particle temperature of 1198 K is achieved for the 20mm diameter combustion chamber and it occurs outside the barrel. After reaching the peak temperature, the particle experiences some temperature drop. Figure 8 shows the average gas temperature and particle temperature at the at the barrel exit. Here, the decreasing trend can be clearly observed. 6 Copyright 2013 by ICAE2013

7 The gas temperature drops from 1343 K to 1310 K whereas the particle temperature drops from 1158 K to 1133 K. Similarly, the gas phase velocity also remained unaffected by the increase in combustion chamber diameter. A very small drop is observed in the gas velocity owing to the increase in residence time of the gas within the combustion chamber. Since, the particles follow the gas phase; no effect is observed when the chamber diameter is varied. The peak velocity of 568 m/s is achieved by the titanium particle, outside of the spray system. Figure 8 shows the variation of averaged gas and particle velocities at the barrel exit. The results show that the average gas and particle phase velocities decrease as the diameter is increased. This is because as the diameter is increased, the chamber volume increases and as a result, the residence time for the gas in the chamber increases resulting in the loss of velocity. The drop in the gas phase velocity is from 1141 m/s to 1122 m/s whereas the drop in the particle velocity is from 569 m/s to 563 m/s. 4.2 Effect of combustion chamber length The combustion chamber length also affects the gas phase velocity and temperature which in turn affect the particle velocity and temperature. The chamber length is varied from 30mm to 90mm. The results obtained are depicted in Figures The results show that as the combustion chamber length increases, the gas phase and particle temperatures decrease. This is due to the increase in the area for heat transfer. More heat is transferred in the longer combustion chamber which reduces the gas phase temperature. The temperature sensitive titanium particle attains peak temperature of around 1200K at the outside of the barrel. The gas phase and particle velocities from the barrel entrance are plotted in Figure 9. The results show that as the combustion chamber length increases, both gas phase and particle velocities decrease. The gas velocity decreases since the gas has to travel a little longer as the length of the chamber is increased. During this travel, the gas loses some velocity magnitude in the combustion chamber itself due to friction with the walls of the chamber. Therefore, a higher gas velocity is achieved by using a shorter combustion chamber. The particle velocity decreases as the chamber length increases. The particle velocity has no significant effect in the barrel for all the cases but as soon as it leaves the barrel, the effect begins to appear. 4.3 Effect of mixing chamber length The mixing chamber length affects the gas and particle phases, the results of which are shown in Figures The range of chamber length investigated is from 30mm to 60mm. The results show that the gas phase in the barrel is greatly affected by the chamber length. As the mixing chamber length is increased, the gas phase temperature de Figure 7 Effect of combustion chamber diameter on gas and particle phase along the center line Figure 8 Effect of combustion chamber diameter on temperature and velocity of the gas and particle phase at the exit creases. This is due to an increase in the gas phase residence time in the mixing chamber and consequently higher cooling before entering the converging diverging nozzle. The particle temperature also decreases as the chamber length increases with more profound effect on the outside of the barrel. The effect is more clearly depicted in Figure 12 showing the average gas and particle phase temperatures at the barrel exit. Unlike the temperature, the gas phase velocity is not much affected by the mixing chamber length. The effect is 7 Copyright 2013 by ICAE2013

8 Figure 9 Effect of combustion chamber length on gas and particle phase along the center line Figure 11 Effect of mixing chamber length on gas and particle phase along the center line Figure 10 Effect of combustion chamber length on temperature and velocity of the gas and particle phase at the exit visible only inside the barrel where the entrance velocity for a 30mm length chamber is 1895m/s as opposed to 1471m/s for a 60mm length chamber. The particle velocity profiles in the barrel overlap indicating that the mixing chamber length does not have significant effect on the particle velocity inside the barrel. A closer look at the exit velocities shows that the average gas and particle velocities at the barrel exit also decrease as the mixing chamber length is increased because of the additional frictional losses which occur in the Figure 12 Effect of mixing chamber length temperature and velocity of the gas and particle phase at the exit longer mixing chambers. 4.4 Effect of diverging diameter of the convergingdiverging nozzle The exit diameter of the diverging section of the converging diverging nozzle is varied from 10mm to 22mm and the results obtained are shown in Figures For this particular case, the results are plotted from the start of the combustion chamber to visualize the effect in the 8 Copyright 2013 by ICAE2013

9 converging diverging nozzle. The effect on gas velocity can be seen from the start of the diverging section. As expected, the velocity is higher for larger exit diameter and it decreases as the diameter is decreased. Also, the increase in diameter makes the shocks move inside of the gun and consequently, the drop in velocity is higher. The particle velocity is also greatly affected. As the diameter is increased, the particle velocity decreases and the effect is observed right from the particle injection location. The difference in velocity is not much for the diameters of 10mm and 14mm but there is a relatively lager difference in particle velocities between 14mm and 18mm diameter nozzle. This is due to the fact that the shocks for 14mm diameter nozzle occur near the exit of the barrel. However, for the 18mm diameter nozzle, the shocks occur at the entrance of the barrel resulting in a great reduction in the gas phase velocity throughout the barrel. The difference of particle velocities between 18mm and 22mm diameter nozzle is not so high. The exit gas phase and particle velocities decrease as the diameter of the diverging section increases. The particle velocity at the exit of the barrel for the lowest exit diameter nozzle is about 568m/s whereas for the largest exit diameter, it is about 281m/s as shown in Figure 14. Since the particle follows the gas phase, this large difference is expected. The converging diverging nozzle has effect on the velocity which in turn affects the temperature. The gas temperature remains almost constant till the end of the mixing chamber. After mixing with the coolant and due to an increase in diameter, the gas phase temperature decreases in the converging diverging nozzle. The maximum temperature attained by the gas is for the largest diameter which is around 2800K but reduces rapidly on the outside of the barrel. Particle temperature is also greatly affected by the change in diameter. As the diameter is increased from 10mm to 14mm, the particle temperature decreased. However, beyond 14mm, the particle temperature increases with an increase of the diameter. The effect of varying diverging section diameter on the exit gas phase and particle temperatures is shown in Figure 14. The gas temperature increases as the diameter is increased and the change is very large when compared to other geometrical parameters. The particle temperature also first decreases as the diameter is increased from 10mm to 14mm and then increases for the other two cases. The change in particle temperature is also significant (~900K). 5. CONCLUSION The effect of several geometrical parameters of a dual stage thermal spray gun on its performance has been studied. The parameters are the length and diameter of the combustion chamber, converging nozzle, mixing chamber and diverging section of the nozzle. The results of the inves Figure 13 Effect of diverging section diameter on gas and particle phase along the center line Figure 14 Effect of diverging section diameter on temperature and velocity of the gas and particle phase at the exit tigations identify the important parameters which affect the titanium particle temperature and velocity. The results show that an increase in the combustion chamber diameter decreases both the particle temperature and velocity, whereas the combustion chamber length has more profound affect than its diameter. Mixing chamber diameter also does not have significant effect. On the other hand, mixing chamber length does affect the particle temperature right from the particle injection location. The diverging section diameter of the converging diverging 9 Copyright 2013 by ICAE2013

10 nozzle is found to be the most sensitive parameter as it has significant effects on the particle temperature and velocity. Generally, as the diameter increases, particle temperature increases whereas particle velocity decreases. REFERENCE [1] Boyer RR. An overview on the use of titanium in the aerospace industry. Materials Science and Engineering 1996; 213: [2] Tabbara H, Gu S, McCartney DG. Computational modeling of titanium particles in warm spray. Computers and Fluids 2011;44: [3] Pawlowski L. The science and engineering of thermal spray coatings. 2nd ed. England: Wiley; [4] Brunette DM, Tengvall P, Textor M, Thomsen P. Titanium in medicine: material science, surface science, engineering, biological responses and medical applications. Berlin: Springer; [5] Donachie Jr MJ. Titanium: a technical guide. 2nd ed. OH: ASM International; [6] Pohler OEM, Unalloyed titanium for implants in bone surgery. Injury 2000;31:7 13. [7] Ponticaud C, Grimaud A, Denoirjean A, Lefort P, Fauchais P. Thermal Spray 2001 New Surfaces for a New Millennium: Proceedings of the International Thermal Spray Conferences. 2001; Singapore. [8] Wu T, Kuroda S, Kawakitam J, Katanoda H, Reed R. Processing and properties of titanium produced by warm spraying. in Proceedings of the 2006 international thermal spray conference. 2006; Seattle, Washington, USA. [9] Papyrin A, Kosarev V, Klinkov S, Alkhimov A, Fomin VM. Cold spray technology. Amsterdam: Elsevier; [10] Browning J. Thermal spray method and apparatus for optimizing flame jet temperature. US patent 5, 330, 798; [11] Kawakita J, Kuroda S, Krebs S, Katanoda H. In situ Densification of Ti Coatings by the Warm Spray (Two Stage HVOF). Process Material Transaction 2006;47: [12] Kawakita J, Kuroda S, Fukushima T, Katanoda H, Matsuo K, Fukanuma H. Dense titanium coatings by modified HVOF spraying. Surface and Coatings Technology 2006;201: [13] Kawakita J, Katanoda H, Watanabe M, Yokoyama K, Kuroda S. Warm Spraying: An improved spray process to deposit novel coatings. Surface and Coatings Technology, 2008;202: [14] Kuroda S, Kawakita J, Watanabe M, Katanoda H. Warm Spraying A Novel Coating Process Based on High Velocity Impact of Solid Particles. Science Technology, Advanced Materials 2008;9:1 17. [15] Cheng D, Xu Q, Tapaga G, Lavernia E. A numerical study of high velocity oxygen fuel thermal spraying process. Part I: Gas phase dynamics. Metallurgical and Materials Transactions A 2001;32: [16] Gu S, Eastwick C, Simmons K, McCartney D. Computational fluid dynamic modeling of gas flow characteristics in a high velocity oxy fuel thermal spray system. Journal of Thermal Spray Technology 2001;10: [17] Kamnis S, Gu S. 3 D modeling of kerosene fuelled HVOF thermal spray gun. Chemical Engineering Science 2006;61: [18] Katanoda H, Kiriaki T, Tachibanaki T, Kawakita J, Kuroda S, Fukuhara M. Mathematical Modeling and Experimental Validation of the Warm Spray (Two Stage HVOF) Process. Journal of Thermal Spray Technology 2009;18: [19] Katanoda H, Morita H, Komatsu M, Kuroda S. Experimental and numerical evaluation of the performance of supersonic two stage high velocity oxy fuel thermal spray (Warm Spray) gun. Journal of Thermal Science 2011;20: [20] Dolatabadi A, Mostaghimi J, Pershin V. Effect of a cylindrical shroud on particle conditions in high velocity oxyfuel (HVOF) spray process. Journal of Materials Processing Technology 2003;137: [21] Shamim T, Xia C, Mohanty P. Modeling and analysis of combustion assisted thermal spray processes. International Journal of Thermal Sciences 2007;46: [22] Cheng D, Xu Q, Lavernia E, Tapaga G. The effect of particle size and morphology on the in flight behavior of particles during high velocity oxy fuel thermal spraying. Metallurgical and Materials Transactions B 2001;32: [23] Li M, Christofides PD. Multi scale modeling and analysis of an industrial HVOF thermal spray process. Chemical Engineering Science 2005;60: [24] Li M, Shi D, Christofides PD. Diamond Jet Hybrid HVOF Thermal Spray Gas Phase and Particle Behavior Modeling and Feedback Control Design. Ind. Eng. Chem. Res 2004;43: [25] Gordon S, McBride BJ. Computer program for calculation of complex chemical equilibrium compositions and applications, in NASA Reference Publication : Lewis Research Center, Cleveland, OH, USA. [26] Ranz WE, Marshall WR. Evaporation from drops (part I). Chem Eng Prog 1952;48: Copyright 2013 by ICAE2013