Search

This paper is devoted to the experimental study of the characteristics of a continuous optical discharge (COD) sustained by high power continuous wave laser radiation at a wavelength λ = 1.08 μ m in high pressure argon. New data on the COD threshold laser power dependence of argon pressure in the range 20–50 bar is obtained. The COD threshold laser power is shown to be in good agreement with the data obtained by other authors and theoretical evaluations provided the contribution of plasma energy loss due to thermal radiation is taken into account properly. The maximum plasma temperature was estimated to be 20–21 kk or higher, favorable to obtain high UV spectral radiance. A study of the convective plume oscillations around COD in argon has been carried out. It is found that in the pressure range 25–35 bars the growth of the laser radiation power leads to a decrease in convection oscillation frequency from 33 to 29 Hz, while the radius of the convective plume grows accordingly. The oscillation frequency ν and characteristic radius of the convective plume r 0 were found to obey the similarity relation ν = 0.5 g / 2 r 0 previously established in experiments with COD in xenon. These results are promising for using COD in argon as a high brightness broadband UV radiation source. [ABSTRACT FROM AUTHOR]

This work investigates the transient anode temperature and plasma parameters considering the electrode movement and the arc expansion. A transient self-consistent model is established using the dynamic mesh approach. Two cases with different maximum arc currents, representing various anode activity, have been studied. The simulations predict significant changes in spatial distribution of species densities, electron temperature and anode surface temperature in case with the anode spot formation. Furthermore, the influence of opening velocity on the plasma parameters has been studied numerically. The model was validated by comparison with experimental data for anode surface temperature as well as spectrally filtered arc images. Results of this comparison are presented and discussed. [ABSTRACT FROM AUTHOR]

Computational fluid dynamics plasma simulations of a capillary capacitively coupled radio frequency thruster discharge with front powered electrode was performed. Good agreement with experimental results was obtained for the global parameters of the thruster. The spatial characteristics of the discharge were analyzed. DC bias is found to develop in accordance with classical theory of asymmetric capacitive discharge. The electrons temperature in the plasma bulk is ∼2 eV and increasing the sheath near the powered electrode to ∼6 eV. The plasma density is 1019 m−3 whereas excited argon species are 1020 m−3. The thrust can be completely attributed to gas dynamic acceleration. [ABSTRACT FROM AUTHOR]

In this work, high-voltage pulsed Ar gas–liquid discharge synergizing iron-based catalyst-activated persulfate (PS) was employed to degrade methylene blue (MB) in water. The catalytic performances of two types of iron-based catalysts, namely the homogeneous catalyst Fe2+ and the heterogeneous catalyst nano-Fe3O4, were compared. Correspondingly, the plasma gas temperature and excited species were calculated and diagnosed using optical emission spectra. It was found that the introduced plasma process significantly enhanced the degradation efficiency of MB by the PS/Fe2+ and the PS/Fe3O4 systems. After 20 min of treatment, the MB degradation efficiency reaches 97.5% and 83.1% in the hybrid plasma/PS/Fe2+ and plasma/PS/Fe3O4 systems, respectively, which is 37.9% and 35.6% higher than that in the PS/Fe2+ and PS/Fe3O4 systems. The synergistic mechanism and key reactive species responsible for MB degradation in hybrid plasma/PS/Fe2+ and plasma/PS/Fe3O4 were explored using the addition of radical scavengers and control experiments under various conditions. The homogeneous catalyst Fe2+ exhibits better activation performance in PS and plasma than that of the heterogeneous catalyst nano-Fe3O4. [ABSTRACT FROM AUTHOR]

OH density and rotational temperature were measured in a saturated water vapor slot-excited microwave plasma using spatio-temporally resolved laser-induced fluorescence. The microwave power was 20–100 W under continuous wave mode, whereas 40–200 W peak under pulse-modulated mode (30% duty cycle, 100 Hz). The water vapor pressure was 2.3 kPa. An approximately 2 mm thick flat-shaped plasma was generated on the surface of a slot antenna. The OH density and rotational temperature in the plasma ranged from 1 × 1014 cm−3 to 1 × 1015 cm−3 and from 1100 K to 3000 K, respectively. OH was produced via various routes originating from electron collisions (e + H2O) and was primarily lost through recombination (OH+OH → H2O+O) and diffusion. The production rate (kp) of OH per H2O molecule, which was calculated based on a simplified reaction model using the measured OH density and temperature, showed a relation of , where P represents the microwave power. This relationship implies a two-step production process of OH via the vibrational excitation of H2O. Although the OH density increased with increasing microwave power, the density became saturated at high power values. [ABSTRACT FROM AUTHOR]

The properties (densities of electrons and metastable argon atoms, effective electron temperature and dust charge) of argon/dust and pure argon pulsed plasmas are studied using a spatially-averaged model. The calculated time-dependencies for the densities of electrons and metastable atoms are compared with the experimental measurements and are found to be in a good qualitative agreement. It is analyzed how the plasma properties depend on the shape of the electron energy probability function (EEPF), the pulsing frequency and the duty cycle for both dust-free and dusty plasma. The analysis reveals that the agreement between theory and experiment is better with Druyvesteyn EEPF than the Maxwellian EEPF. Further, the variation in the pulsing frequency νp differently affects the metastable density nm in a dust-free and in a dusty plasma. For large νp, the metastable density in the dust-free pulsed plasma is larger than in the continuous-wave (CW) discharge, while the opposite is obtained in the presence of dust particles. This difference probably arises because of faster variation in the effective electron temperature in the dusty plasma due to collection of electrons by dust particles. Our calculations also show that dust particles may affect the behavior of electron density in the beginning of the on-period due to an enhancement in electron collection by dust particles. [ABSTRACT FROM AUTHOR]

The existence of low-frequency disturbance can affect the discharge stability of a hollow cathode, and may even lead to the extinction of the discharge. In this paper, a semi-transient algorithm based on the classical 0-D theoretical model of an orificed hollow cathode was developed to study the discharge stability of a hollow cathode under a continuous low-frequency electric field disturbance. A laboratory experiment using an orificed BaO-W cathode was carried out to confirm the validity of transient analysis results and to determine an appropriate time step in the analysis. The theoretical evolution trend of the discharge current is largely in agreement with the experiment for a time step value of 12.5 μs although there is a 7% error in the discharge duration prediction. This study has identified the role of two emission enhancement mechanisms in the failure process of a self-sustained discharge, namely the Schottky effect and the sheath funneling effect. The presence of the sheath funneling effect induces larger amplitude oscillations and a sharp drop in discharge parameters when the effect ceases to exist, and this significantly weakens the discharge stability. The influence of operating conditions and disturbance parameters on transient discharge characteristics was examined for the NASA NSTAR cathode. Within the parameter range that was studied, findings show that higher xenon flow rates or lower discharge currents provide optimal duration as a result of higher electron temperatures and denser plasma at the initial state. For a fixed operation point, the discharge duration was found to decline exponentially with disturbance amplitude and quadratically with disturbance frequency. [ABSTRACT FROM AUTHOR]

The volume of material removed from the workpiece by a pulse (Vv) in the electric discharge machining was quantitatively determined using a multi-physics simulation model in ANSYS CFX software. Electrical heating is used in the model to simulate the plasma channel formation by defining the boundary and time-controlled current density initial conditions. Time-dependent physical properties at plasma temperature were used to reflect the actual processing environment. The heat was transferred from the plasma channel to the workpiece by electrical heating from the electrode, and Vv was calculated by means of the amount of heat transfer. The calculated Vv values for AISI4140, Ti6Al4V and Inconel 718 workpieces were lower than the experimental results and the difference was observed to change between 38.3% and 46.9%. [ABSTRACT FROM AUTHOR]

In this article, we discuss the electron density and the temperature of plasmas in solution, which produce an attractive reaction field for processing. The measurement of the electron density and the estimation of the electron temperature of the plasma in NaCl solution were performed using spatiotemporally resolved optical emission spectroscopy. The electron densities were estimated to be of the order of from the Stark broadening of the Hβ line. The electron temperatures of the plasma were estimated to be in the range of 0.2–1.4 eV based on the electron-neutral bremsstrahlung. The estimated values were confirmed to be consistent with the experimental values of voltage and current density in the assumption of the Boltzmann equation. [ABSTRACT FROM AUTHOR]

Plasma temperatures of free-burning arcs dominated by Cu vapour are presented. It is shown that with increasing optical thickness the methods for optical thin evaluation of spectral lines fail and even methods for temperature determination in optically thick plasmas are limited. It reveals that a strongly simplified radiation model leads to more realistic plasma temperatures in the investigated range of plasma parameters. Temperatures between 10 000 and 20 000 K and Cu partial pressures between 0.1 and 0.5 bar are found at currents between 1 and 4 kA. Corresponding current densities range from 35 to 166 A mm−2. [ABSTRACT FROM AUTHOR]

Presently, most high intensity discharge (HID) lamps contain mercury to generate a high pressure buffer gas and thereby an appropriate power input into the arc. Due to its toxicity, the replacement of Hg is of particular interest in recent research on HID lamps. Up to now, the emission coefficient of an atomic Hg double line is widely used to determine the plasma temperature Tpl in HID lamps. Tpl is needed to calculate the total density of atoms and ions of elements inside these lamps. A combination of optical emission and broadband absorption= spectroscopy allows us to evaluate Tpl independently of Hg emission lines. The method is required for a determination of Tpl if the Hg line intensity within the investigated lamp is too low, is superimposed by other lines or if environmental-friendly Hg-free lamps are developed. Within this work, phase-resolved plasma temperatures are determined in front of the electrode of Hg-containing MH lamps by emission spectroscopy at atomic Hg lines. Above all, temperatures are measured by a combination of emission and absorption spectroscopy at atomic rare earth lines, namely Dy and Tm. A comparison of Tpl determined by both methods agree within an error margin of <10%. Total phase-resolved rare earth atom densities are obtained by means of the measured ground state densities and Tpl. The combination of emission and absorption spectroscopy is also applied to the bulk plasma of lamps where the intensity of the Hg emission lines is too low for plasma temperature measurements or Hg is absent. It provides the partial rare earth pressure and by comparison with thermodynamic data cold spot temperatures within the lamps. [ABSTRACT FROM AUTHOR]

In this paper we introduce an experimental technique that allows for high-speed, three-dimensional determination of electron density and temperature in axially symmetric free-burning arcs. Optical filters with narrow spectral bands of 487.5-488.5 nm and 689-699 nm are utilized to gain two-dimensional spectral information of a free-burning argon tungsten inert gas arc. A setup of mirrors allows one to image identical arc sections of the two spectral bands onto a single camera chip. Two-different Abel inversion algorithms have been developed to reconstruct the original radial distribution of emission coefficients detected with each spectral window and to confirm the results. With the assumption of local thermodynamic equilibrium we calculate emission coefficients as a function of temperature by application of the Saha equation, the ideal gas law, the quasineutral gas condition and the NIST compilation of spectral lines. Ratios of calculated emission coefficients are compared with measured ones yielding local plasma temperatures. In the case of axial symmetry the three-dimensional plasma temperature distributions have been determined at dc currents of 100, 125, 150 and 200A yielding temperatures up to 20000K in the hot cathode region. These measurements have been validated by four different techniques utilizing a high-resolution spectrometer at different positions in the plasma. Plasma temperatures show good agreement throughout the different methods. Additionally spatially resolved transient plasma temperatures have been measured of a dc pulsed process employing a high-speed frame rate of 33000 frames per second showing the modulation of the arc isothermals with time and providing information about the sensitivity of the experimental approach. [ABSTRACT FROM AUTHOR]

The thermal state of a glow discharge with intermediate current in flowing atmospheric air is investigated by a combination of Rayleigh scattering thermometry imaging and numerical simulation. Results from the simulation indicate that during the initial breakdown the local translational temperature can reach a huge value (e.g. 6000 K) but decreases quickly due to strong heat transfer to the surrounding cold air. In the gliding stage, the translational temperature of plasma is balanced by the input power density and the heat dissipation rate. As the gas flow rate is increased, the translational temperature in the glow plasma column diminishes. The flow affects the thermal state of plasma from two aspects. First, it promotes elongation of the plasma column to decrease the input power density. Second, the flow enhances local heat dissipation. As a result, the translational temperature is lowered due to flow. Using a two-temperature model, which considers the translational temperature, the vibrational temperature and their transitions, the non-thermal state of plasma is further analyzed. The gas flow is found to reduce the translational temperature and the vibrational–translational relaxation rate, and thus prevent thermalization of the plasma column. [ABSTRACT FROM AUTHOR]

The emission spectra from an industrial, droplet-based, laser-produced plasma, extreme ultraviolet light source for nanolithography are here presented and analyzed. The dependence of spectral features on the CO2-drive-laser intensity is studied by changing the beam spot size at constant pulse energy and duration. We characterize the spectrum by fitting the results of atomic structure calculations to the short-wavelength region (7–11 nm), where the contributions from various charge states can be resolved, and obtain the relative contributions of charge states Sn9+ –Sn15+ . These relative contributions are compared to charge state populations as calculated with the non-equilibrium plasma kinetics code flychk. The calculations are shown to be in good qualitative agreement with the results, showing that the effective plasma temperature, and with it, the shape of the unresolved, main emission feature at 13.5 nm, is a remarkably weak function of laser intensity under this source normal operating conditions. [ABSTRACT FROM AUTHOR]

A model for arc-cathode attachment in gas metal arc welding is presented. It considers the cathodic heating in the case of a non-refractory iron cathode, with the assumption of the lowering of the local plasma temperature, due to cold metal vapor. It takes the plasma bulk temperature as well as the cathode drop voltage as free parameters and it gives a maximum of heat flux and current density for a cathode surface temperature below boiling temperature. The applicability as a heat source description in a weld pool simulation has been shown, and the temperature field of the cathode was calculated, giving rise to heat flux and current density distributions, which differ significantly from the converntionally used axisymmetric Gaussian distribution. The maximum cathode surface temperature was lower than boiling temperature, which is in agreement with the observeration. [ABSTRACT FROM AUTHOR]

We investigate the temperature-dependence of the light emission properties of Cu-dominated plasmas with pressures in the order of 1 bar and for currents between 0 and 2 kA. While the wavelength of Cu I 3d94s5s transitions for currents above 1 kA is essentially constant, a 0.1 nm Stark blueshift is observed with decreasing current from 1 kA to zero. Using a lineshape analysis of the light emission spectra, which is based on solving a one-dimensional radiative transfer equation along the line-of-sight across the arc, we show that the transition wavelength increases linearly with temperatures in the 5000–10 000 K range. Based on this observation, we propose a simple method for the temperature diagnostic of plasmas that neither requires an absolute calibration of the intensity of the detected signal, nor complex simulation tools. [ABSTRACT FROM AUTHOR]

A plasma injector element was designed to experimentally study the mechanism of methane-air diffusive flame stabilized by a discharge plasma. The air plasma was generated within the annulus gap of the injector by alternating current dielectric barrier discharge. The discharge voltage, current and photographs were recorded first. Three internal effects of the plasma on combustion were later investigated separately through several diagnostic methods, including optical emission spectrometry (OES), infrared thermography, thermocouple, infrared thermometer, schlieren imaging, photos and CH* chemiluminescence. Finally, the return on investment (ROI) was calculated. The results showed that a large number of filamentary micro discharge paths occur within the discharge gap. These discharge paths rotate anticlockwise at high speed and act as a virtual ‘fan’ to induce the flow jet. The velocity of the induced jet increases with increasing discharge voltage. The original jet expansion angle is enlarged by the radial velocity component of the induced jet, resulting in the mixing enhancement of the air and methane. The plasma rotational temperature (the first negative system ) obtained from OES is close to the discharge gas temperature measured by infrared thermography, indicating that the discharge gas temperature can be approximately represented by the rotational temperature. According to the measured temperature of the injector and the jet, the impact of the thermal effect of the plasma on flame stabilization is negligible. Due to the plasma, the height of the flame center and its representative length are generally reduced as the voltage rises, and the methane-air mixture becomes ignitable, and a stable flame can be reached under the conditions in which direct ignition fails. The combustion is enhanced with increasing heat release rate of the flame by the plasma. This finding revealed that the ROI of plasma-assisted flame stabilization is lower under a higher flowrate and a larger equivalence ratio for unstable flame situations. [ABSTRACT FROM AUTHOR]

The study of gas–metal arc welding (GMAW) often relies on the hypothesis of local thermodynamical equilibrium (LTE). Few works are dedicated to its validation and none is focused on the case of pulsed metal active gas welding. This paper presents the result of a spectroscopic diagnostic performed on a pulsed welding arc with 82%Ar–18%CO2 shielding gas. High-speed imaging and electrical data recording are also performed. The excitation temperature is obtained with the Boltzmann plot method, while iron and argon line Stark broadening measurement is used to obtain the electron temperature and density. The iron mole fraction in the plasma is calculated by solving the conservation equations. Partial LTE hypothesis validity across the arc is discussed considering the agreement between the two temperatures, electron density and iron content. The results show supporting evidence for the main part of the plasma, along radial and axial directions. This can be associated with high iron vapour concentration, close to 100% near the consumable anode tip. Discrepancies occur only at the fringe of the arc, where the two temperatures differ by more than 2000 K. The temperature drop close to the arc axis observable for the upper part is weaker when compared with an inert gas welding operation. [ABSTRACT FROM AUTHOR]

SiOCH ULK films with k-value from 2.5 to 2.1 and porosity from 24 to 40% were etched in CHF3, CHF3  +  Ar, CF4 and CF4  +  Ar plasmas at  +15…−120 °C with and without bias being applied. It was shown that the presence of Ar in gas mixture can significantly increase the damage of unetched ultra low-k (ULK) material (at sidewalls) due to the removal of  −CH3 groups from the film by VUV photons. It was also shown that etching and damage of the sidewalls by F atoms can be partially prevented by lowering the temperature of the sample. [ABSTRACT FROM AUTHOR]

We generated H2O-ice dielectric barrier discharge (DBD), which has H2O ice as a dielectric barrier, at a wide range of cryogenic temperatures down to 6.5 K, and controlled plasma gas temperature as a control parameter. H2O-ice DBD exhibited variations in optical emissions and transition of discharge modes with changing plasma gas temperature. Furthermore, H2O-ice DBD is revealed to be anticipated to have sufficient reactivity as a reaction field on the ice surface in comparison with ultraviolet irradiated chemical reactions such as formation of amino acids. This study is the first report on gas temperature-controlled plasma generation coexistent with H2O ice, focusing on plasma gas temperature as a control parameter. [ABSTRACT FROM AUTHOR]

This review describes the history and development of plasma-assisted catalysis focussing mainly on the use of atmospheric pressure, non-thermal plasma. It identifies the various interactions between the plasma and the catalyst that can modify and activate the catalytic surface and also describes how the catalyst affects the properties of the discharge. Techniques for in situ diagnostics of species adsorbed onto the surface and present in the gas-phase over a range of timescales are described. The effect of temperature on plasma–catalysis can assist in determining differences between thermal catalysis and plasma-activated catalysis and focuses on the meaning of temperature in a system involving non-equilibrium plasma. It can also help to develop an understanding of the gas-phase and surface mechanism of the plasma–catalysis at a molecular level. Our current state of knowledge and ignorance is highlighted and future directions suggested. [ABSTRACT FROM AUTHOR]

Potential determinants of the laser-induced breakdown spectra in N2–O2 mixtures are investigated with varying species concentration, laser energy, and exposure time (equal to the detector’s intensifier gate time). A focused laser beam with pulse energy 29 mJ, 38 mJ, or 46 mJ generates the breakdown plasma in a N2 mixture containing O2 from 0% to 21% in mole fraction. Time-resolved measurements of the breakdown emission spectra with 2 or 10 ns exposure are carried out beginning from the arrival of the laser beam at the focal point up to several μs later. These measurements show the temporal evolution of the instantaneous spectrum in the range 550–825 nm that contains multiple N and O atomic/ionic emission lines. Analyses of the baseline-subtracted spectra, normalized by the peak intensity of the N atomic emission line at 747 nm, indicate that the temporal evolution of each emission line is independent of species concentration and laser energy, though the instantaneous emission line intensity is strongly dependent on the species concentration and exposure time. The plasma temperature, estimated using two atomic oxygen lines and assuming local thermal equilibrium, and its temporal evolution are shown to be independent of the species concentration and a weak function of the laser energy at a given gas density, e.g. atmospheric pressure and room temperature. In addition, thermal equilibrium calculations, for predicting atomic and ionic species concentrations in the plasma, are carried out, and the combined results dictate that the emission-line intensity is linear with the atomic/ionic species concentration because of the consistent plasma temperature and its temporal evolution, independent of species concentration. [ABSTRACT FROM AUTHOR]

In this paper, a novel method of the dual-wavelength (laser-induced breakdown spectroscopy LIBS) technique using a single laser system is proposed and demonstrated. Experiments are performed using a pulsed Nd3+ : YAG laser with a pair of 355–1064 nm and also with 532–1064 nm. The shorter wavelength laser is used for ablation and plasma formation, and the fundamental wavelength (1064 nm) is used for plasma re-excitation. The proposed dual-wavelength LIBS technique is used for lunar simulant samples under different ambient pressure conditions. Various characteristic parameters, such as the emission line-intensity enhancement, plasma temperature, lifetime and plasma area, are studied. Experimental studies clearly showed the emission line-intensity enhancement up to a factor of 3. Emission lifetime showed a longer sustained emission with an increase of up to 33% for the dual-wavelength approach. A theoretical simulation based on the hydrodynamic equations is also performed for dual-wavelength ablation and re-excitation. The estimated plasma temperature and ablation plume-front velocity clearly showed an increase in dual wavelength, which is in agreement with the experimental results. [ABSTRACT FROM AUTHOR]

The calculated values of thermodynamic and transport properties of pure F2 and fluorocarbon compounds CF4, C2F2, C2F4, C2F6, C3F6 and C3F8 at high temperatures are presented in this paper. The thermodynamic properties are determined by the method of Gibbs free energy minimization, using standard thermodynamic tables. The transport properties, including electron diffusion coefficients, viscosity, thermal conductivity and electrical conductivity, are evaluated using the Chapman–Enskog method expanded up to the third-order approximation (second order for viscosity). The most accurate cross-section data that could be located are used to evaluate collision integrals. The calculations based on the assumption of local thermodynamic equilibrium are performed for atmospheric-pressure plasmas in the temperature range from 300 to 30 000 K for different pressures between 0.1 and 10 atm. The results of F2, CF4, C2F2, C2F4 and C2F6 are compared with those of previously published studies. Larger discrepancies occur for transport coefficients; these are explained in terms of the different values of the collision integrals that were used. The results presented here are expected to be more accurate because of the improved collision integrals employed. [ABSTRACT FROM AUTHOR]

The motion of an evaporated atom of the cathode material in a near-cathode plasma is considered. It is shown that the evaporated atom is ionized almost instantly. The created ion, under the influence of a strong electric field existing in the cathode proximity, has a high probability of returning to the cathode. A small fraction of evaporated atoms are able to diffuse away from the cathode to the region where they are involved in plasma flow and lose their chance to return to the cathode. The fraction of the total evaporated atoms, which do not return to the cathode, the escape factor, determines the net erosion rate. In order to calculate this factor, the distributions of the plasma parameters in the near-cathode plasma were considered. Calculations showed that the escape factor is on the order of a few per cent. Using experimental data on the plasma and cathode temperatures, we calculated the net erosion rate for a free-burning 200 A argon arc with a thoriated tungsten cathode. The calculated erosion rate is close to 1 µg s−1, which is in agreement with available experimental data. [ABSTRACT FROM AUTHOR]

The density of OH radicals in ground state is measured by laser-induced fluorescence (LIF) spectroscopy in the core of a micro-flow discharge in He, Ar and N2 with a water electrode. The lines P2(6), P1(4) and P2(3) of the X2Π(v″ = 0) to the A2∑+ (v′ = 1) transition are used for OH radical excitation. The density of the main quencher of OH radical in the core of the discharge is estimated based on the time decay of the LIF signal. It is revealed that the plasma core consists of a high amount of 8–10% of water vapour. The calculation of the absolute density of OH radical is carried out based on the model of LIF excitation including vibrational and translation energy transfer, and the results in different gases are presented for the discharge. [ABSTRACT FROM AUTHOR]

The controlled metal transfer process (CMT) is a variation of the gas metal arc welding (GMAW) process which periodically varies wire feeding speed. Using a short-arc burning phase to melt the wire tip before the short circuit, heat input to the workpiece is reduced. Using a steel wire and a steel workpiece, iron vapour is produced in the arc, its maximum concentration lying centrally. The interaction of metal vapour and welding gas considerably impacts the arc profile and, consequently, the heat transfer to the weldpool. Optical emission spectroscopy has been applied to determine the radial profiles of the plasma temperature and iron vapour concentration, as well as their temporal behaviour in the arc period for different mixtures of Ar, O2 and CO2 as shielding gases. Both the absolute iron vapour density and the temporal expansion of the iron core differ considerably for the gases Ar + 8%O2, Ar + 18% CO2 and 100% CO2 respectively. Pronounced minimum in the radial temperature profile is found in the arc centre in gas mixtures with high Ar content under the presence of metal vapour. This minimum disappears in pure CO2 gas. Consequently, the temperature and electrical and thermal conductivity in the arc when CO2 is used as a shielding gas are considerably lower. [ABSTRACT FROM AUTHOR]