Parametric study of the decomposition of NH3 for an induction plasma reactor design

The present paper reports a study of the gas mixing and chemical transformation in an induction plasma reactor under atmospheric pressure, and its dependence on the plasma operating conditions. For this purpose, the thermal dissociation of ammonia into nitrogen and hydrogen was chosen because of the relative simplicity of the reactions involved and its use in a number of studies on plasma synthesis of ultrafine nitride ceramic powders using ammonia as nitriding agent. A hot-wall reactor configuration is investigated in which ammonia is injected radially through multiple orifices into the gases at the exit nozzle of an induction plasma torch. Concentration mapping in the mixing zone was carried out, using a VG-Micromass-PC 300 D quadrupole mass spectrometer, for different plasma power levels, in the range 13–24 kW. A 3-point injection mode is used with the injection ports oriented upstream at 45° to the torch axis. This allows uniform mixing of the injected gas in the plasma jet. A systematic study of the effects of plate power and ammonia and plasma gas flow rates on the mixing and dissociation of NH₃ in the reactor is reported. The results are analyzed and discussed from the viewpoint of their use for optimizing the design of induction plasma reactors, to he applied to the vapor-phase synthesis of ultrafine silicon nitride powders.     

Mixing study of the induction plasma reactor: Part I. Axial injection mode

A systematic study of the gas-mixing pattern in an induction plasma reactor under atmospheric and low pressure conditions is reported. Different reactor configurations were investigated in which nitrogen is injected as an auxiliary gas either axially into an Ar/H₂, discharge in the center of the induction coil region, or radially through multiple orifices, into the plasma jet at tire exit nozzle of the torch. Concentration mapping in the mixing zone was carried out, using a VG-Microniass-PC 300 D mass spectrometer at plasma power levels and reactor pressures, in the range of 13–24 k 6V and 35–93 kPa, respectively. Comparison of these results with cold-flow measurements underlined the substantial difference in the mixing pattern in each of these two cases. A considerably faster mixing of the gases is noted under cold flow conditions compared to that in the presence of the discharge. The results are discussed from the viewpoint of their use for optimum reactor design applied to tire vapor-phase synthesis of ultrafine ceramic powders, using induction plasma technology.     

Influence of the Shroud Gas Injection Configuration on the Characteristics of a DC Non-transferred Arc Plasma Torch

The characteristics of the plasma jet emanating from a dc non-transferred plasma torch is affected by many factors including arc current, type of gas, gas flow rate, gas injection configuration and torch geometry. The present work focuses on experimental investigation of the influence of shroud gas injection configuration on the I–V characteristics and electro-thermal efficiency of a dc non-transferred plasma torch operated in nitrogen at atmospheric pressure. The plasma gas is injected into the torch axially and shroud gas is injected through three different nozzles such as normal, sheath and twisted nozzles. The effects of flow rates of plasma/axial gas and arc current on I–V characteristics and electro-thermal efficiency of the torch holding different nozzles are investigated. The I–V characteristics and electro-thermal efficiency of the torch are found to be strongly influenced by the shroud gas injection configuration. The effect of arc current on arc voltage decreases with increasing the axial gas flow rate. At higher axial gas flow rate (>?45 lpm), the I–V characteristics of the plasma torch are similar irrespective of the nozzle used. The variation of electro-thermal efficiency with arc current is almost similar to that of arc voltage with arc current. As expected, the electro-thermal efficiency is increased when the axial gas flow rate is increased and at higher axial gas flow rate, it is not influenced by the arc current and shroud gas configuration. The plasma torch with normal nozzle may be better in the range of operating conditions used in this study.     

Measurements of Temperatures and Electron Number Density in an Argon–Nitrogen Plasma Jet Generated by a dc Torch-Operation Close to Supersonic Threshold

An investigation of the plasma jet generated by a dc argon–nitrogen plasma torch, operated in association with a controlled-pressure chamber, is presented. The purpose of this article is to describe a study of the properties of a subsonic plasma jet under such operating conditions, when its transition to supersonic flow regime is nearly complete. The goal is that of performing plasma diagnostics not only in the initial region of the jet but also in the downstream region where the plasma emission is weak. For this purpose two different diagnostic methods are used. The first approach is based on non-intrusive optical emission spectroscopy, which yields both excitation and rotational temperatures as well as electron number density fields. The zone investigated by this method extends from the torch exit to about 10 nozzle diameters downstream. The second approach consisted of the use of the intrusive enthalpy probe technique for the measurement of the plasma gas temperature, mainly in the tail region of the plasma jet. In the present work, the effects of axial and radial distances across the jet, on the temperature and electron density profiles are discussed for subsonic flow conditions. Interesting features revealed are the data shown for the various diagnostic methods, which either disagree or overlap with each other. Finally, our results show the need for involving non-equilibrium models for the argon–nitrogen plasma due to the presence of significant differences between the temperatures of light and heavy particles.     

Generation of Long, Laminar Plasma Jets at Atmospheric Pressure and Effects of Flow Turbulence

Long, laminar plasma jets at atmospheric pressure of pure argon and a mixture of argon and nitrogen with jet length up to 45 times its diameter could be generated with a DC arc torch by restricting the movement of arc root in the torch channel. Effects of torch structure, gas feeding, and characteristics of power supply on the length of plasma jets were experimentally examined. Plasma jets of considerable length and excellent stability could be obtained by regulating the generating parameters, including arc channel geometry, gas flow rate, and feeding methods, etc. Influence of flow turbulence at the torch nozzle exit on the temperature distribution of plasma jets was numerically simulated. The analysis indicated that laminar flow plasma with very low initial turbulent kinetic energy will produce a long jet with low axial temperature gradient. This kind of long laminar plasma jet could greatly improve the controllability for materials processing, compared with a short turbulent arc jet.     

Laser Doppler anemometry under plasma conditions. Part I. Measurements in a d.c. plasma jet

A study is presented on the use of laser Doppler anemometry (LDA) techniques for the measurement of the gas and particle velocities under plasma conditions. Experimental data is presented for a d.c. plasma jet in which alumina particles are injected under different operating conditions. The results reveal that the plasma velocity at the exit of the jet is of the order of 200–300 m/s. The intensity of turbulence is as high as 30 to 40% in the free shear layer and the particle velocity distribution is shown to be asymmetric, with particle dispersion in the plane of injection considerably more important than that in the perpendicular direction. The average particle velocity depends on the composition of the plasma gas, the torch current, and power.     

Characterization of the Converging Jet Region in a Triple Torch Plasma Reactor

Enthalpy probe measurements were taken of the converging plasma plume in a triple torch plasma reactor and related to substrate heat flux measurements. Results show excellent entrainment of process gases injected into the converging plasma plume by way of the central injection probe. At lower pressures (40 kPa), the plasma volume is equivalent to at least a 3 cm diameter, 4 cm long cylinder, with relatively uniform temperature, velocity, and substrate heat flux profiles when compared to a typical dc arc jet. Converging plasma plume size, substrate heat flux, and enthalpy profiles are also shown to be a strong function of applied system power. Substrate heat flux measurements show smaller radial gradients than enthalpy probe measurements, because of the high radial velocity component of gases above the substrate boundary layer. Enthalpy probe measurements were also conducted for diamond deposition conditions and approximate temperature and velocity profiles obtained. Problems with the uniform gas mixture assumption prohibited more accurate measurements. Reproducibility of enthalpy measurement results was shown with an average standard deviation of 11.8% for the velocity and 7.6% for the temperature measurements.     

Nozzle optimization for dissociated species transport in low pressure plasma chemical vapor deposition

Numerical simulation has been used to study the fluid dynamics and chemical kinetics in a supersonic nozzle situated downstream of a plasma reactor. To assist in system scale-up and optimization, effective nozzle design can help in maximizing the transport of chemically active species to the substrate. This paper examines the chemical non-equilibrium of the flow, the effect of different flow parameters, and the effect of different nozzle geometries. A three species hydrogen-argon gas mixture was modeled with finite dissociation and recombination. Non-equilibrium transport and thermodynamic mixture properties based on species pair collision cross sections were implemented. Running a plasma torch off design power can significantly alter power losses through the nozzle wall and subsequently change the species distribution at the nozzle exit. Decreasing the effective nozzle throat diameter can notably decrease atomic concentrations. Small changes in upstream or downstream pressures have a negligible effect on supersonic species transport through the nozzle. Flow separation can be avoided by correctly designing the divergent portion of the nozzle.     

Velocity Measurements for Arc Jets Produced by a DC Plasma Spray Torch

An optical method was used to determine the axial velocity of plasma jets produced by a DC plasma spray torch. Different experimental conditions were tested in order to systematically study the influence of the working parameters on the plasma velocity. In this way, the arc current ranged between 200 and 600 A, the gas flow rate between 30 and 80 slm, and the internal nozzle diameter between 6 and 10 mm; the plasma gases were either an Ar–H ₂ mixture or N ₂ . Rather well defined tendencies were observed and at the same time it appeared that the arc stability greatly influenced the fluctuations of the velocity.     

Preparation of Silicon Nanopowder by Recycling Silicon Wafer Waste in Radio-Frequency Thermal Plasma Process

Silicon nanopowders were prepared from silicon waste by using radio-frequency thermal plasma. Silicon waste, generated from the manufacturing process of silicon wafers, was pulverized to form micrometer-sized silicon starting powder. In order to obtain as much silicon nanopowder as possible from thermal plasma processing, the enhancement of vaporization and the quenching rate of the silicon starting powder were considered as major factors. A counter-flow injection apparatus (CFIA) was introduced for improved vaporization and homogeneous nanoparticles. It was designed to inject argon as a quenching gas in the direction opposite the thermal plasma flame flow. The controlled location of the CFIA injection nozzle and the flow rate of the quenching gas affect the residence time of the injected staring powder by recirculating flow and the vapor density by gas mixing. The variation of the flow pattern inside the reactor and the characteristics of the products were investigated to determine the optimal processing environment to prepare uniform and small silicon nanopowder particles. The environment was defined by two parameters: the flow rate of the counter quenching gas and the distance between the torch and CFIA nozzles. The flow rate of the quenching gas was controlled from 30 to 70 L/min. The distance between the torch and CFIA nozzles was adjusted from 150 to 350 mm. When the quenching gas flow rate of 70 L/min and the distance of 350 mm were applied, the uniform and smallest silicon nanopowders were obtained.     

Measurement of Temperature in the Steam Arcjet During Plasma Arc Cutting

Spectroscopic measurements were performed by observing the plasma inside the kerf during cutting of stainless steel using direct current electric arc. Experiments were carried out on the plasma torch operated with the plasma gas composed of the vaporized mixture of water and ethanol; arc current was 60 A and cutting speed 30 cm/min. Emission spectral lines of neutral iron were used to experimental evaluation of the temperature of plasma in the kerf and close under the cut plate. Complicated nature of the plasma inside the kerf, including presence of metallic vapours and departures from equilibrium, was taken into account. Hence relatively reliable results were obtained, from which it was possible to get insight into the energy balance and cutting performance of the torch. Temperature of the plasma in the kerf was substantially lower than at the nozzle exit of the torch; however the temperature drop along the kerf was small.     

Modeling of plasma jets with superimposed vortex flow

This work is concerned with analytical studies of thermal plasma jets, which are finding increasing interest for thermal plasma processing. A two-dimensional model for turbulent plasma jets with superimposed vortex flow has been developed, incorporating multiple time scales for velocity and temperature fluctuations and a density-weighted averaging for the density fluctuation effect. Results show that adding swirl to the flow field for confined and free jets induces strong axial and radial pressure gradients near the nozzle exit, causing a rapid decay of the axial velocity with increasing distance from the nozzle. Comparisons with cold flow show similar trends close to the nozzle exit, but further downstream, the axial velocities increase again, especially for larges swirl numbers. Comparisons of theoretical predictions based on the present model with available experimental data are, in general, in reasonable agreement.     

New design of a radio-frequency plasma torch

A numerical model has been developed for predicting the two-dimensional flow and temperature fields in a radio-frequency (rf) plasma torch. The method employed here is based on Boulos' model with the exception of the boundary conditions for the electric and magnetic field equations. Calculations have been made for the confirmation of a new sample injection method, which is capable of completely evaporating refractory materials at high feeding rates without interfering with the stability of the plasma. In the newly designed torch, the reagent is radially injected into the hottest part of the plasma through quartz capillary tubes set symmetrically between an inductor coil. Experimental investigations have also been performed for verifying the proper function of the design. These results provide evidence that our radial injection method developed here is more effective in practical processing than the conventional axial injection methods.     

Comparative Study of Flow Characteristics Inside Plasma Torch with Different Nozzle Configurations

A numerical analysis of the influence of different nozzle configurations on the plasma flow characteristics inside D.C plasma torches is presented to provide an advanced nozzle design basis for plasma spraying torches. The assumption of steady-state, axis-symmetric, local thermodynamic equilibrium, and optically thin plasma is adopted in a two-dimensional modeling of plasma flow inside the plasma torch. The PHOENICS software is used for solving the governing equations, i.e. the conservation equations of mass, momentum, and energy along with the equations describing the K-epsilon model of turbulence. The calculated arc voltages are consistent with the experimental results when arc current, gas inflow rate, and working gas are the same as the experimental parameters. Temperature, axial velocity contours inside plasma torches, profiles along the torch axis and profiles at the outlet section are presented to show the plasma flow characteristics. Comparisons are made among those torches. The results show that torches with different anode nozzle configurations produce different characteristics of plasma flows, which suggest some important ideas for the advanced nozzle design for plasma spraying. In order to validate the model and to show its level of predictivity, a comparison of the model with experimental results encountered in the literature is presented in the last part.     

Synthesis of Ozone at Atmospheric Pressure by a Quenched Induction-Coupled Plasma Torch

The technical feasibility of using an induction-coupled plasma (ICP) torch to synthesize ozone at atmospheric pressure is explored. Ozone concentrations up to ~250 ppm were achieved using a thermal plasma reactor system based on an ICP torch operating at 2.5 MHz and ~11 kVA with an argon/oxygen mixture as the plasma-forming gas. The corresponding production rate and yield were ~20 g ozone/hr and ~2g ozone/kWh, respectively. A gaseous oxygen quench formed ozone by rapid mixing of molecular oxygen with atomic oxygen produced by the torch. The ozone concentration in the reaction chamber was measured by Fourier Transform infrared (FTIR) spectroscopy over a wide range of experimental conditions and configurations. The geometry of the quench gas flow, the quench flow velocity, and the quench flow rate played important roles in determining the ozone concentration. The ozone concentration was sensitive to the torch RF power, but was insensitive to the torch gas flow rates. These observations are interpreted within the framework of a simple model of ozone synthesis.     

Characterization of d.c. plasma torch voltage fluctuations

The arc root fluctuations at the anode-nozzle of a d.c. plasma spray torch with a special configuration of the electrodes allowing to work with the same gas flowrate with nozzle diameters ranging from 6 to 10 mm were systematically studied. The plasma gas was Ar/H2 (25 vol % H2), the current was varied between 200 and 600 A and the plasma gas flowrate between 24 and 80 slm. After 30–60 mn working the nozzle wall started to be sufficiently eroded to have a stagnant arc spot which lived until arcing created another one. It was shown that the life time of the upstream arc spots were 30–40 % longer than the downstream ones which could play an important role in the electrode erosion. Dimensional analysis allowed to find a relationship between the nozzle diameter D, the arc current I and gas flow rate G and the mean spot lifetime which is closely connected with the difference between D and the electrical diameter of the arc column. The comparison of voltage signal and light emission at a point of the plasma jet close to the nozzle exit on its axis allowed to determine the mean electrical field within the plasma column and the mean position of the arc root. The comparison with the electrode erosion area for well defined conditions showed a good correlation with the calculated arc root position.     

Parametric study of the flow and temperature fields in an inductively coupled r.f. plasma torch

A theoretical investigation of the effect of different parameters on the flow and the temperature fields in a radiofrequency inductively coupled plasma is carried out. The parameters studied are: central injection gas flow rate, total gas flow rate, input power, and the type of plasma gas. The results obtained for argon and nitrogen plasmas at atmospheric pressure indicate that the flow and the temperature fields in the coil region, as well as the heat flux to the wall of the plasma confinement tube, are considerably altered by the changes in the torch operating conditions.     

Preliminary spectroscopic experiments with helium microwave induced plasma produced in air by use of a new structure: the axial injection torch

In this experimental study we have used spectroscopic methods to characterize helium plasma obtained by means of a novel waveguide-fed microwave plasma torch at atmospheric pressure, the axial injection torch. This device produces a “plasma flame” by coupling high frequency (HF) power at 2.45 GHz to the discharge. Various flame parameters (namely the electron density number and the electron and gas temperatures) have been determined by using spectroscopic diagnostic techniques that provided an estimate in terms of the helium flow rate, absorbed HF power and axial position in the experiments. These preliminary results suggest some departure from local thermodynamic equilibrium (LTE) and seem to indicate the utility of the discharge as an excitation source for emission spectroscopy. Comparison with other microwave torches already described in the literature is made in terms of the electron density and the electron and gas temperature.     

An Abel inversion method for radially resolved measurements in the axial injection torch

A method for the radial analysis of the plasma torch produced by the axial injection torch device at atmospheric pressure is proposed. The method uses a fast high-resolution acquisition set-up including an intensified charge coupled device and an image rotation system, as well as a new data processing procedure based on the Abel inversion technique that provides the radial profile of thin plasmas. Use of the Fernández-Palop smoothing technique significantly reduces distortion in the data profile. The accuracy of the proposed method was assessed by applying it to the estimation of the populations of HeI excited states.     

Particle velocity measurements in induction plasma spraying

Laser Dopple anemometry (LDA) measurements of the particle velocity are carried out during an induction plasma spraying operation. The velocity of nickel alloy particles, or molten droplets, at the exit of an induction plasma torch prior to impact on the substrate is shown to vary with the plasma and powder injection conditions. Plasma spraying under soft vacuum (150–450 Torr) gives rise to substantially higher particle velocities (40–60 m/sec) compared to those attained at atmospheric pressure (10–20 m/sec).