Skip to content
BY-NC-ND 3.0 license Open Access Published by De Gruyter September 15, 2015

Optimal Design of Nozzle for Supersonic Atmosphere Plasma Spraying

  • Pei Wei EMAIL logo , Zhengying Wei , Guangxi Zhao , Y. Bai and Chao Tan

Abstract

Through numerical simulation, key issues concerning the plasma jet features as well as the sizes of nozzle for supersonic atmosphere plasma spraying (SAPS) were analyzed in this paper. Numerical results were compared with the experimental measurements and a good agreement has been achieved. Due to the effect of mechanical compression, the increasing sizes of r1, r2, r3 and r4 (r1, r2, r3 and r4 are the sizes of nozzle) lead to a decrease in temperature and velocity of plasma jet. But large size of r5 can increase the external temperature and velocity of plasma jet, which benefit particles accelerating at the far downstream region. A new nozzle was designed based on the simulation results. Compared to the temperature and velocity of plasma jet in the original nozzle, the maximum temperature and velocity of plasma jet in new structure are increased by about 9.8% and 44.5%, which is a benefit to the particles to reach a higher speed and surface temperature.

Introduction

Thermal spraying is a group of processes in which the metallic or ceramic materials are deposited in a molten or semi-molten state on a prepared substrate. The process involves the formation of combustion or plasma flame, and particle deposition on the prepared substrate. Recently, an advanced high-efficiency supersonic atmosphere plasma spraying system (SAPS) has been successfully developed by national key laboratory for remanufacturing (China) for the deposition of ceramic and metallic coatings with good performance while the energy consumption was greatly reduced (lower than 60 KW) as compared with “PlazJet” [15]. The high-energy plasma jet results from the structural design of SAPS gun with a Laval nozzle. For the specific structure, the plasma shows unusual properties such as high temperature, up to 20,000 K, high velocity, up to 2,400 m/s, steep temperature and velocity radial gradients [6]. The main advantage of SAPS is that there is no limitation to the melting temperature of the sprayed materials and good quality in coating. This significantly extends the technology possibility to any material that could be melted, including refractory ceramics. SAPS is highly efficient, reliable and easy to handle.

During SAPS, working gases are injected into the nozzle via gas vortex ring in order to pinch the arc and increase the life of plasma gun. Owing to the special layout of the inner nozzle contour a long quasi-laminar plasma jet is produced that reduces the reaction between powder and surrounding residual oxygen. Plasmas are quasi-neutral multiparticle systems characterized by gaseous and mixtures of free electrons and ions, as well as neutral particles (atoms, molecules, radicals) with a high mean kinetic energy of electrons or all plasma components, and a considerable interaction of charge carriers with the properties of the system. Reactive plasma-forming gases, including dissociation, ionization and recombination reactions, influence the energy transport, and the presence of multiple species determining the values of the transport coefficients. Specific properties of plasma different from the neutral gases rely on its response to electromagnetic forces [7]. Plasma can be produced by transferring energy into a gas until the energy level is sufficient to ionize the gas, allowing the electrons and ions to act independently of one another. SAPS have very high temperatures and energy densities and are consequently surrounded by very steep gradients of temperatures and densities [8]. Kinetic and composition non-equilibrium are usually the consequence, and description of the energy transport requires consideration of these effects. In particular in the electrode regions, the gradients can be so steep (in the order of 108 K/m or several hundred degrees per mean free path) that the continuum concept may break down and discontinuities in properties may be encountered. Heat transfer from arc plasma is characterized by several distinct features such as transport of dissociation and ionization energy and of electrical charges in addition to mass transport [9].

The structure of spraying gun has significant influence on the temperature and velocity of plasma jet so as to effect on the particle flight status. The interaction between the particle and the plasma jet is a critical factor determining the final coating properties, which to a large extent depend on the plasma jet temperature and velocity. To understand the effect of structure of plasma gun on plasma jet behavior so as to improve the coating quality, an accurate description of transport phenomena of plasma jet is essential. With the rapid development of computational fluid dynamic (CFD) methods, some researchers started to study the characteristics of plasma jet by numerical methods [1013]. In plasma density computation method, B. Liu et al. [10] use the results of Chen et al. [14] as comparison and conclude that there is no much difference between ideal gas law and Maxwell energy distribution law to determine plasma gas density. Erick Meillot et al. [15] adopted a Joule heating model to simplify the arc heating zone in the plasma gun. Many workers have found that plasma torch inside the gun demonstrates three-dimensional (3-D) properties [11, 16], to take a better knowledge of the plasma flow three-dimensional modeling must be done. To minimize the boundary conditions effects, both the cathode and the anode solid should be taken into consideration [17].

In this paper, a 3-D computational model was built to study the effects of structure of SAPS gun on temperature and velocity of plasma jet. In order to verify the validity of the calculation model, the velocity and temperature of the flight zirconia particles were measured by SprayWatch 2i, and the heating and melting processes of zirconia particle in the supersonic plasma spraying were analyzed using previous CFD methods. The numerical results agreed well with the experimental data. The optimal sizes of spraying gun were calculated by the 3-D CFD model.

Numerical model

Thermodynamic and transport properties

For supersonic plasma, the numerical simulation needs to take various physical and chemical phenomena into consideration. Chemical reactions such as ionization, dissociation, recombination and so on taking place in the plasma jet can be treated using a general kinetic algorithm. The plasma-forming gas is Ar + H2 in this paper. So the following species were considered: e, Ar, Ar+, ArH+, H, H+, H2+ and Ar2+, and their related excited atoms [18, 19]. The chemical reaction j can be symbolized by

(1)iaijXiibijXi,

where Xi represents species i, aij and bij are stoichiometric coefficients for reaction j. The thermodynamic and transport properties of given plasma mixture are functions of the degree of ionization, which is usually obtained by using the Saha equation:

(2)nr+1,inenr,i=2zr+1zr2πmekBTreac3/2h3expEr,ikBTreac,

where nr,i and nr+1,i are the number density of ionization of rth and r+ 1th ionized state for the element i. zr and zr+1 are the partition functions for the consecutive state of ionization r and r+ 1, which further depend upon temperature and pressure. me represents the electron mass. h and kB are Planck constant number and Boltzmann constant number, respectively. Er,i represents the ionization energy of reaction. Treac represents the temperature of ionization reaction. The population of different states of ionization of various elements comprising the plasma mixture is calculated by the above equation.

In thermal plasmas, the number of elastic collisions between electrons and heavy species is sufficient to reach local thermal equilibrium (LTE) when electron number density is high [20]. The LTE conditions only reached in the core of dc plasma jets or arcs at atmospheric pressure. While in the arc or jet fringes, the electron number density overcome the heavy species dramatically, and the electrons kinetic temperature Te is higher than that of heavy species Th. A two-temperature model is applied to take account of this kind of thermal nonequilibrium [21]. In this paper, the kinetic model is applied to obtain plasma compositions. This model, provided all reactions and coefficients are known, gives realistic compositions. The rate of chemical reaction in the forward or reverse direction can be expressed in terms of forward and reverse reaction rate coefficients, Rf and Rr, respectively. The rate of coefficients can be obtained from an Arrhenius-type relation [22]. If we consider the reaction present as A+B+CD+E:

(3)Rf=aTbexpcT,
(4)Rr=Rf2πmAmBmCkTmDmEh23/2gAgBgCgDgEexpEkBT,

where a, b and c are constants that are usually obtained from experimental data, g and m are statistical weight and mass of the species, respectively, involved in the reaction evaluated at temperature T, and h is Planck constant. Sometimes the forward rate coefficients can be calculated using the collision cross-sectional data through the following equation:

(5)Rf=2eme0αfεσεεdε,

where fε is the electron energy distribution function (Maxwellian distribution), σε is the collision cross section and e, me, ε are the electron charge, electron mass and energy, respectively. In calculation of the particle concentration, several chemical reactions of Ar and H have been considered, and the constants a, b and c for Arrhenius type have been given experimentally [20, 22], and the reaction energy data for collision cross-sectional data were also given experimentally [22, 23].

Plasma-governing equations

In numerical model, the equations governing homogenous gas flow express the following several parts: continuity, momentum and thermal energy equations for the multicomponent fluid mixture, species equations for each component of the mixture. In gas heated by arc, a dynamic equilibrium exists between dissociation, ionization energy and recombination, and the thermodynamic properties depend on the plasma composition. All the model of the plasma therefore must contain not only the conservation of mass, momentum and energy but also Maxwell’s equation and current conservation. The continuity equation can be represented as ρg/t+ρgVg=0, where ρg is gas density, Vg is gas velocity vector. Plasma gas is a dense cloud of electrons, ions, atoms and molecules. In numerical model, the species continuity equation can be represented as [24]

(6)nit+niVi=dωidt

The subscript i represents different species, ni is the ith species particle number per unit volume in m3, Vi is the ith species particles velocity, dωidt is the ith species particles net change rate due to chemical reaction.

The momentum conservation equation links the flow velocity of a fluid element with the external forces acting on it. The following equation is a general one, as the gravity force can be ignored, the last term can be omitted:

(7)ρdVgdt=FνcPg23(μVg)+2μ[ε˙],

where the left term ρdVgdt is the inertia force term, the right term Fνc is volume force term which is equal to the total force of mass force, electric field force and Lorentz force. Pg is the gas pressure gradient. 23(μVg) is viscous stress causing the gas volume expansion, μ is the gas dynamic viscosity. 2μ[ε˙] is the viscous stress caused by fluid deformation when fluid moves, [ε˙] is strain rate tensor. Due to charged particles movement in plasma, the energy equation should contain joule heat generation rate term caused by electromagnetic field EJ:

(8)ρgddtε+Vg22=(PVg)+ρgVg+EJq,

where ε is internal energy, q is the heat through unit area in unit time.

The standard kε model has been adopted in the numerical simulations of turbulent jet under plasma conditions [25, 26]. Although the results predicted by the kε model are not satisfactory, it still provided semiquantitative information on the plasma jet.

The current density in the arc column of a typical high-intensity arc may reach values in excess of 106–1010 A/m2. Arcs may attach to the electrodes, and in particular to the cathode, the current continuity equation is governed by [24]

(9)ρet+J=0,

where ρe is the charge density and J is the electric current density.

The control equation of DC plasma jet in steady condition can be expressed as

(10)1μ2A=σeϕJs,

where A is the magnetic vector potential, σe is the electrical conductivity, ϕ is the electric potential and Js is the other source term. The Lorentz force exists in the plasma jet produced by electromagnetic field. The Lorentz force J×B can add to the magnetohydrodynamic equation as a source term. In the equation current density can be expressed as

(11)J=σeE+ue×B.

The E is the electric field intensity, ue is the velocity of charged particles and B is the magnetic induction intensity. So the Lorentz force FL can be derived by the following formula:

(12)FL=σeϕ+u×B×B.

The coupling relationship between electric field and magnetic field can be solved by Maxwell’s equations:

(13)×E=Bt,
(14)×B=μ0J,
(15)B=0.

In this paper, a 3-D numerical model for plasma jet is developed to investigate the plasma jet temperature and velocity field. The governing equations for plasma jet consist of the continuity, momentum and thermal energy equations for multicomponent fluid mixture, species equations for each component of the mixture gas, state relations for an ideal gas mixture with temperature-dependent specific heat, enthalpy and transport properties and so on.

Computational model and boundary

The structure of SAPS nozzle and several main sizes are shown in Figure 1. According to the nozzle, the 3-D CFD modeling and mesh generation is shown in Figure 2. There are 5,77,440 nodes and 4,99,567 elements in the whole computational domain. In the radial direction, the grid is more refined near the central axis and coarser toward the outer environment. In the numerical simulations, the boundary conditions are listed in Table 1.

Figure 1: Two-dimensional structure of SAPS nozzle.
Figure 1:

Two-dimensional structure of SAPS nozzle.

Figure 2: Three-dimensional CFD modeling and mesh generation.
Figure 2:

Three-dimensional CFD modeling and mesh generation.

Table 1:

Boundary conditions of geometric model.

BCTemperature/KSpeed/(m/s)Potential ϕ/(A/m2)Magnetic A
Main gas inlet1,000VAr = 65,VH2 = 30ϕ/n=0A/n=0
Carrier gas inlet50030φ/n=0A/n=0
Cathode3,500j=7.02×106A/n=0
Anode1,000ϕ/n=0A/n=0
Exit5000ϕ/n=0A/n=0

The calculations are initialized from the cold gas and then mixed with the incoming jet until the steady state is reached. The steady-state solution is expected to be unique and independent of the initial conditions.

Experiment

Plasma temperature can be measured assuming local thermodynamic equilibrium using atomic or molecular spectroscopy or measured by two-color pyrometry at two wavelengths [2729]. The plasma velocity can be measured with either a Pilot tube probe technique or with a time-of-flight two-point laser anemometer [3033]. But the above methods were difficulty in implementing in SAPS system due to extreme high temperature. So the numerical model provides a useful tool to understanding the plasma jet characteristics in supersonic plasma spraying process.

The numerical model is proved accurately and good to estimate the temperature and velocity of plasma jet and in-flight particles through the experiment. A commercially available monitoring system SprayWatch 2i (Osier, Finland) is used to measure the velocity and surface temperature of the in-flight particles, seen in Figure 3. The fast CCD camera of SprayWatch 2i was triggered by the velocity signal. The CCD focal length is 185 mm and the exposure time is 5 μm. The signal is real time (about few hundreds of nanoseconds) handled thanks to a controller program and this procedure allows both obtaining the particle velocity (with an accuracy of about 10%) and triggering cameras. In addition, the temperature of in-flight particles was measured by an optical sensor head and a fast two-color pyrometer. SprayWatch 2i system adopts two optical with double-color filter which can penetrate different wavelength (700 and 850 nm). Each signal was transmitted through a bandpass filter with wavelength and then detected using an avalanche silicon photodetector. The ratio of the radiation intensity at these wavelength was used to calculate the particle temperature with an accuracy of ±100C. The in-flight particles surface temperature can expressed as

(16)E(λ,T)=ε(λ,T)C1πλ5expC2λT11,
(17)T=C21λ21λ1lnE(λ1)E(λ2)5lnλ2λ11,

where ε is the radiation coefficient of particle, E is the radiation energy in J, λ is the flame wavelength in nm, λ1 and λ2 are the selected wavelengths in nm, C1 and C2 are constants.

Figure 3: Diagnosis system of temperature and velocity of in-flight particles for SAPS.
Figure 3:

Diagnosis system of temperature and velocity of in-flight particles for SAPS.

Results and discussion

Model validation

When the particles fly through the plasma flame, the mass, momentum and energy transfer from the flame will dramatically change the particle properties. The Lagrangian method is used to trace the in-flight particles individually. Particle acceleration and heating are tracked after injection, by considering the drag force by plasma jet and heat transport on the particle surface. A one-dimensional model for particle heating and melting is used in which a spherical shape of the particle is assumed and internal convection within the molten part of the particle is neglected. In this section, the main process parameters and the nozzle sizes are listed in Tables 2 and 3, respectively (Figure 4).

Table 2:

The main process parameters of supersonic plasma spray.

ParametersValue
Working gasAr-H2(H2-20%)
Flow of main gas70 L/min
Pressure of main gas0.8 MPa
Powder feed gas flow7 L/min
Current420 A
Power65 kW
Powder feed rate45 g/min
Table 3:

The supersonic plasma spraying gun main dimension.

Main dimensionValue
d114 mm
d26.8 mm
r14.0 mm
r24.0 mm
r34.9 mm
r45.6 mm
r55.6 mm
L18.5 mm
L215 mm
α85°
Figure 4: Velocity and temperature distributions of particles.
Figure 4:

Velocity and temperature distributions of particles.

In the simulation, the initial particle size is 50 μm. The velocity and temperature distribution of particles at axial direction along the center line of plasma jet versus spraying distance for experiment and simulation are shown in Figure 4. It is clearly seen that the computational temperature and velocity of particle is slightly higher than the measurement (Figure 5). The maximum error of temperature and velocity between experiment and simulation were 1.7% and 3.3%, respectively (error = (Ss– Se)/Ss, Ss and Se are the result of simulation and experiment, respectively). The neglects of particle breakage and evaporation effect on the particles temperature and velocity bring error into the experimental and numerical results. In general, the experimental and simulation results show that, within the error allowed, the calculation model is considered to estimate the temperature and velocity of in-flight particles and plasma jet accurately.

Figure 5: Temperature (a) and velocity (b) distribution of particle at axial direction along the center line of plasma jet versus spaying distance for experiment and simulation.
Figure 5:

Temperature (a) and velocity (b) distribution of particle at axial direction along the center line of plasma jet versus spaying distance for experiment and simulation.

Effect of the throat size

The throat sizes of nozzle include r1, r2 and r3 as shown in Figure 1. Based on the sizes listed in Table 3, selected sizes of r1 are 3, 4, 5, 5.8 and 6.5 mm in the simulation to research the influence of r1 on the temperature and velocity of plasma jet, and the numerical results are shown in Figure 6. In order to observe the variation regularity of temperature and velocity of plasma jet clearly, the temperature and velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance are shown in Figure 7.

Figure 6: Temperature distribution (a) and velocity distribution (b) of plasma jet for nozzle with various r1 sizes.
Figure 6:

Temperature distribution (a) and velocity distribution (b) of plasma jet for nozzle with various r1 sizes.

Figure 7: Influence of r1 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 7:

Influence of r1 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

Due to the increasing mechanical compression with the decreased r1, the density of plasma becomes larger, which usually cause increasing in maximum temperature and velocity of plasma jet (shown in Figure 8). But the temperature and velocity outside the nozzle were nearly identical. The temperature of plasma jet is decreased by 4.82% when the r1 increase from 3 to 6.5 mm (temperature decreases from 16,400 to 15,610 K), as well as the velocity of plasma jet is decreased by 8.83% (velocity decreases from 1,993 to 1,817 m/s). In order to improve the gun efficiency, the r1 should be select as small as possible.

Figure 8: Maximum temperature (a) and maximum velocity (b) of plasma jet decrease with the increasing size of r1.
Figure 8:

Maximum temperature (a) and maximum velocity (b) of plasma jet decrease with the increasing size of r1.

The effect of r2 on the temperature and velocity of plasma jet are shown in Figure 9 (r2 = 3, 4, 5.4, respectively). Obviously the influence of r2 on the maximum temperature of plasma jet is slight and only decreased by 3.02% when the r2 increased from 3 to 5.4 mm (temperature decreases from 16,200 to 15,710 K). But the maximum velocity of plasma jet has significant decrease with the increasing r2. The velocity is decreased by about 33.7% when the r2 increases from 3 to 5.4 mm (velocity decreases from 2,765 to 1,834 m/s). As r1 the r2 should be selected as small as possible to improve the gun efficiency.

Figure 9: Effect of r2 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 9:

Effect of r2 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

Finally the effect of r3 on the temperature and velocity of plasma jet are shown in Figure 10 (r3 = 3, 4, 4.9, 5.5 mm, respectively). It is clearly shown that r3 has the same trend of r1 and r2 influence on the maximum temperature and velocity of plasma jet. The temperature and velocity are decreased by about 11% and 50.9% when the r3 increases from 3 to 5.5 mm (maximum temperature decreases from 17,360 to 15,450 K, and maximum velocity decreases from 3,546 to 1,741 m/s). As same as the r1 and r2, the r3 should be selected as small as possible to improve the gun efficiency. But to ensure particles can eject out of spraying gun fluently, the size of r1, r2 and r3 should not be less than 3 mm.

Figure 10: Effect of r3 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 10:

Effect of r3 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

Effect of r4 and r5

The r4 is limited by r1, r2 and r3 and its sizes of 5 and 5.6 mm are selected. The temperature and velocity of plasma jet all decrease with the increasing r4, and the decreasing rate of temperature is lower than that of velocity (Figure 11). The maximum temperature and velocity of plasma jet are reduced about 1.94% and 15.7% when the r4 increases from 5 to 5.6 mm (maximum temperature decreases from 16,020 to 15,710 K, and maximum velocity decreases from 2,175 to 1,834 m/s).

Figure 11: Effect of r4 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 11:

Effect of r4 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

When compared with r4, the internal temperature and velocity of plasma jet almost has no change with the increasing size of r5 (r5 increases from 5.6 to 6.5 mm), as shown in Figure 12. But the external temperature and velocity of plasma jet increase with the increasing size of r5. It has an obvious change on the velocity of plasma jet with the increasing r5. The velocity of plasma jet at the nozzle exit increases to about 49.8% (velocity increases from 875 to 1,311 m/s), which benefits the particles to accelerate in the plasma jet.

Figure 12: Effect of r5 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 12:

Effect of r5 on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

As shown in Figure 13, there are little variation of temperature and velocity of the plasma jet on different α. In fact, the effect of d1 and d2 on the temperature and velocity of the plasma jet as well as α were very small and can be neglected in this paper. Particles with zero and small injection angles (0–10°) at certain carrier gas flow rate tend to enter into high temperature zone, where it is easily heated up to the melting temperature due to strong heat transfer between particles and plasma jet. So the α should be selected appropriately.

Figure 13: Effect of α on the temperature and velocity of the plasma jet.
Figure 13:

Effect of α on the temperature and velocity of the plasma jet.

Optimal design

From the above analysis, the effect of d1, d2 and α on the temperature and velocity of the plasma jet is very small, which can maintain the original size in new structure. But there are obvious influences of r1r5 on temperature and velocity of the plasma jet and the degree to which size effect on the temperature and velocity of the plasma jet is r3 > r2 > r4> r1 > r5. When compared with the original sizes of spraying gun (Table 3), the r3 should firstly selected in new structural design of spraying gun, and then the other sizes are selected on the premise that there is no deformation on the Laval tube of spraying gun. The optimal sizes of nozzle are r1 = 3 mm, r2 = 4 mm, r3 = 4 mm, r4 = 5 mm, r5 = 5.6 mm, respectively. The temperature and velocity distribution of plasma jet computed using the optimal sizes are shown in Figure 14. It is clearly shown that the maximum temperature and velocity of plasma jet in new structure increased to about 9.8% and 44.5% (maximum temperature and velocity of plasma jet are 17,250 K and 2,651 m/s, respectively), which is of benefit to the accelerating particle to reach higher surface temperature and flight speed.

Figure 14: Effect of optimal sizes on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.
Figure 14:

Effect of optimal sizes on (a) temperature distribution and (b) velocity distribution of plasma jet at axial direction along the center line of plasma jet versus spaying distance.

Conclusion

The theoretical analyses have been performed to investigate the influence of sizes of nozzle on the temperature and velocity of plasma jet in SAPS. The numerical model was proved good to estimate the temperature and plasma jet by experiment. Simulation shows that the decreasing r1, r2, r3 and r4 lead to increase in mechanical compression, which usually causes increasing in maximum temperature and velocity of plasma jet. But the increasing size of r5 can improve the external temperature and velocity of plasma jet. In general, the extent of size effect on the temperature and velocity of plasma jet is r3> r2> r4> r1> r5. So in design of new nozzle, the r3 should be determined firstly, and then the other sizes of nozzle are selected on the premise that there is no anamorphic in the Laval tube of nozzle. Compared to temperature and velocity of plasma jet in original nozzle, the maximum temperature and velocity of plasma jet in new structure increased by about 9.8% and 44.5%, which is favorable for particles to accelerate to higher flying speed and surface temperature.

Funding statement: Funding: This work was financially supported by the National Science Foundation of China (grant/award number: 51202187).

Acknowledgments

This work was supported by the State Key Laboratory for Mechanical Behavior materials. Thanks to the help of Supersonic Plasma Spraying Laboratory of Xi’an Jiaotong University.

References

[1] X.C. Zhang, B.S. Xu, Y.X. Wu, F.Z. Xuan and S.T. Tu, App. Surf. Sci., 254 (2008) 3879–3889.10.1016/j.apsusc.2007.12.023Search in Google Scholar

[2] X.C. Zhang, B.S. Xu, S.T. Tu, F.Z. Xuan, H.D. Wang and Y.X. Wu, Appl. Surf. Sci., 254 (2008) 6318–6326.10.1016/j.apsusc.2008.03.148Search in Google Scholar

[3] Z.H. Han, B.S. Xu, H.J. Wang and S.K. Zhou, Surf. Coat. Technol., 201 (2007) 5253–5256.10.1016/j.surfcoat.2006.07.176Search in Google Scholar

[4] Z.J. Yin, S.Y. Tao, X.M. Zhou and C.X. Ding, J. Them. Spray. Technol., 18 (2009) 292–296.10.1007/s11666-009-9297-2Search in Google Scholar

[5] Y. Bai, Z.H. Han, H.Q. Li, C. Xu, Y.L. Xu, C.H. Ding and J.F. Yan, Surf. Coat. Technol., 205 (2011) 3833–3839.10.1016/j.surfcoat.2011.01.056Search in Google Scholar

[6] P. Wei, Z. Wei, S. Li, C. Tan and J. Du, App. Surf. Sci., 321 (2014) 538–547.10.1016/j.apsusc.2014.09.159Search in Google Scholar

[7] S.A. Dembovsky, S.A. Kozyukin and E.A. Chechetkina, Mater. Res. Bull., 17 (1982) 801–807.10.1016/0025-5408(82)90031-9Search in Google Scholar

[8] S. Zhu, B.S. Xu and J.K. Yao, Mater. Sci. Forum., 3981 (2005) 475–479.10.4028/www.scientific.net/MSF.490-491.475Search in Google Scholar

[9] X.C. Zhang, B.S. Xu, F.Z. Xuan, H.D. Wang, Y.X. Wu and S.T. Tu, J. Alloy. Compd., 467 (2009) 501–508.10.1016/j.jallcom.2007.12.083Search in Google Scholar

[10] B. Liu, T. Zhang and D.T. Gawne, Surf. Coat. Technol., 132 (2000) 233–243.10.1016/S0257-8972(00)00847-1Search in Google Scholar

[11] F. Qunbo, W. Lu and W. Fuchi, J. Mater. Process Tech., 166 (2005) 224–229.10.1016/j.jmatprotec.2004.08.022Search in Google Scholar

[12] G. Mariaux and A. Vardelle, Int. J. Them. Sci., 44 (2005) 357–366.10.1016/j.ijthermalsci.2004.07.006Search in Google Scholar

[13] I. Ahmed and T.L. Bergman, J. Them. Spray. Technol., 9 (2000) 215–224.10.1361/105996300770349953Search in Google Scholar

[14] X. Chen, Y.P. Chyou, Y.C. Lee and E. Pfender, Plasma Chem. Plasma Process., 5 (1985) 119–141.10.1007/BF00566210Search in Google Scholar

[15] E. Meillot, S. Vincent, C. Caruyer, J. Caltagirone and D. Damiani, J. Them. Spray. Technol., 18 (2009) 875–886.10.1007/s11666-009-9354-xSearch in Google Scholar

[16] H.P. Li and E. Pfende, J. Them. Spray. Technol., 16 (2007) 245–260.10.1007/s11666-007-9023-xSearch in Google Scholar

[17] R. Bolot, C. Coddet, A. Allimant and D. Billières, J. Them. Spray. Technol., 20 (2011) 21–27.10.1007/s11666-010-9560-6Search in Google Scholar

[18] M. Capitelli and E. Molinari, J. Plasma Phys., 4 (1970) 335–355.10.1017/S0022377800005043Search in Google Scholar

[19] M. Capitelli, A. Laricchiuta, D. Pagano and P. Ttraversa, Chem. Phys. Lett., 379 (2003) 490–494.10.1016/j.cplett.2003.07.013Search in Google Scholar

[20] R.F.G. Meulenbroeks, R.A.H. Engeln, M.N.A. Beurskens, R.M.J. Paffen, M.C.M. van de Sanden and J.A.M. van der Mullen, Plasma Source. Sci., Technol., 4 (1995) 74–85.10.1088/0963-0252/4/1/008Search in Google Scholar

[21] P. Andre´, J. Aubreton, M.F. Elchinger, P. Fauchais and A. Lefort, Plasma Chem. Plasma Process., 21 (2001) 83–105.10.1023/A:1007093412813Search in Google Scholar

[22] M.M. Hossain, Y. Tanaka and T. Sakuta, J. Phys. D Appl. Phys., 35 (2002) 529–535.10.1088/0022-3727/35/6/307Search in Google Scholar

[23] A.V. Phelps, J. Phys. Chem, Ref. Data, 21 (1992) 883–897.10.1063/1.555917Search in Google Scholar

[24] M.I. Boulos, P. Fauchais and E. Pfender, Thermal Plasmas Fundamentals and Applications, Plenum Press, New York (1994).10.1007/978-1-4899-1337-1Search in Google Scholar

[25] C.H. Chang and J.D. Ramshaw, Plasma Chem. Plasma Process., 13 (1993) 189–209.10.1007/BF01466041Search in Google Scholar

[26] K. Cheng and X. Chen, Int. J. Heat Mass Tran., 47 (2004) 5139–5148.10.1016/j.ijheatmasstransfer.2004.06.028Search in Google Scholar

[27] M. Vardelle, A. Vardelle, P. Fauchais and C. Moreau, Meas. Sci. Technol., 5 (1994) 205–212.10.1088/0957-0233/5/3/002Search in Google Scholar

[28] M. Vardelle, A. Vardelle, A.C. Leger, P. Fauchais and D. Gobin, J. Them. Spray. Technol., 4 (1995) 50–58.10.1007/BF02648528Search in Google Scholar

[29] J. Mishin, M. Vardlle, J. Lesinski and P. Fauchais, J. Phys. E Sci. Instrum., 20 (1987) 620–625.10.1088/0022-3735/20/6/008Search in Google Scholar

[30] J. Grey, P.F. Jacob and M.P. Sherman, Rev. Sci. Instr., 33 (1962) 738–741.10.1063/1.1717947Search in Google Scholar

[31] J.B. Cox and F.J. Weinberg, J. Phys. D Appl. Phys., 4 (1971) 877–881.10.1088/0022-3727/4/7/301Search in Google Scholar

[32] G. Gouesbet, Plasma Chem. Plasma Process., 5 (1985) 91–117.10.1007/BF00566209Search in Google Scholar

[33] L.E. Drain, The Laser Doppler Technique, Wiley, New York (1980).Search in Google Scholar

Received: 2015-2-6
Accepted: 2015-7-6
Published Online: 2015-9-15
Published in Print: 2016-8-1

©2017 by De Gruyter

This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Downloaded on 28.3.2024 from https://www.degruyter.com/document/doi/10.1515/htmp-2015-0036/html
Scroll to top button