Table of contents

A plasma jet produced in water using a submerged ac excited electrode in a coaxial dielectric barrier discharge configuration was studied. Plasma jet formation was found to occur only while the source was submerged. Plasma jet operation was characterized with and without gas flow. It was found that over 60% of the discharge power was deposited into the water and did not vary appreciably with excitation frequency. Presumably the remaining power fraction went into excitation, ionization and local electrode heating. Emission spectra of the jet revealed nitrogen, hydrogen, hydroxyl and oxygen emission lines. Operation of the plasma jet in water containing the oxidation–reduction indicator methylene blue dye resulted in a marked clearing of the water as observed visually and with a spectrophotometer, suggesting plasma-induced chemical reactivity.

In this work modifications of the orbital motion limited approach are used to investigate the impact of time varying phenomena on the floating potential of a dust grain. The main interest is focused on different regimes relevant for RF discharges. Three cases are considered. First, the case of the RF sheath. Second, the case of charging in the bulk plasma with a time varying electric field and third, the case when the time varying current in the bulk plasma is carried only by a fraction of the electron population. This last case is relevant to low pressure discharges when stochastic heating is important.

This paper reports a systematic study of spatially extended atmospheric plasma (SEAP) arrays employing many parallel plasma jets packed densely and arranged in an honeycomb configuration. The work is motivated by the challenge of using inherently small atmospheric plasmas to address many large-scale processing applications including plasma medicine. The first part of the study considers a capillary–ring electrode configuration as the elemental jet with which to construct a 2D SEAP array. It is shown that its plasma dynamics is characterized by strong interaction between two plasmas initially generated near the two electrodes. Its plume length increases considerably when the plasma evolves into a high-current continuous mode from the usual bullet mode. Its electron density is estimated to be at the order of 3.7 × 10¹² cm⁻³. The second part of the study considers 2D SEAP arrays constructed from parallelization of identical capillary–ring plasma jets with very high jet density of 0.47–0.6. Strong jet–jet interactions of a 7-jet 2D array are found to depend on the excitation frequency, and are effectively mitigated with the jet-array structure that acts as an effective ballast. The impact range of the reaction chemistry of the array exceeds considerably the cross-sectional dimension of the array itself, and the physical reach of reactive species generated by any single jet exceeds significantly the jet–jet distance. As a result, the jet array can treat a large sample surface without relative sample–array movement. A 37-channel SEAP array is used to indicate the scalability with an impact range of up to 48.6 mm in diameter, a step change in capability from previously reported SEAP arrays. 2D SEAP arrays represent one of few current options as large-scale low-temperature atmospheric plasma technologies with distinct capability of directed delivery of reactive species and effective control of the jet–jet and jet–sample interactions.

Ion energy distribution functions (IEDFs) near a source exit of a magnetically expanding argon plasma using only permanent magnets (PMs) are measured by a combination of a retarding field energy analyzer and a pulsed probe method for three types of magnetic-field configurations and various operating gas pressures, where the magnetic-field strength in the source tube is increased up to 270 G by adding the number of the PM layers. The 13.56 MHz rf power for plasma production is maintained at 250 W. The IEDFs show the existence of an accelerated group of ions. It is found that the energy of the accelerated ions increases when the magnetic-field strength is increased, but saturates at seven times the electron temperature at argon pressures of 0.6–1.6 mTorr. Our results show that the maximum velocity of the accelerated ions is found to be 14 km s⁻¹ with a Mach number of 3.8.

A standard swarm analysis of electron scattering cross sections in nitrous oxide (N₂O) is presented. The experimental results for drift velocities and effective ionization coefficients (differences between the ionization and attachment coefficients), obtained over an extended range of E/N (electric field normalized to the gas number density) by the pulsed-Townsend technique, are compared with the numerical solution of the Boltzmann equation. Our analysis shows that commonly used sets of cross sections have to be modified in order to fit the new experimental data, in particular the dissociative cross sections for attachment and electronic excitation (with a threshold energy of around 4.0 eV). Using a single set of cross sections it was possible to fit both the data for pure N₂O and those for the N₂O/N₂ mixtures with 20%, 40%, 60% and 80% N₂O.

We have diagnosed the electron density in nitrogen plasma generated in a device composed of coaxial-hollow microdielectric-barrier discharges using a millimetre-wave transmission method. At atmospheric pressure the generated plasma occupied almost the entire hollow-electrode space, and the electron density was estimated as 10¹²–10¹³ cm⁻³. This value is in good agreement with theoretical prediction in a simple particle-balance model. We also studied the effect of small admixtures of H₂O and O₂ into nitrogen discharges at atmospheric pressure. In the case of H₂O addition, the measured electron density was lower than that in pure nitrogen, in particular at high reduced electric fields, due to the increase in the loss processes of electrons. In the case of O₂ addition, the electron density decreased with an increase in O₂ concentration due to the additional loss process through the formation of negative ions.

The charge state and energy distributions of ions emitted from a Sn-based laser-produced plasma extreme ultraviolet source, formed at a power density of ≈4 × 10¹¹ W cm⁻², have been recorded as a function of angle. Measurements were recorded from 20° to 80° with respect to the target normal. For each individual ion stage detected, more energetic ions showed preferential emission close to the target normal whereas the less energetic ions exhibited peak emission at larger angles. The peak in the observed ion emission moved to higher energies with increasing ionization at all angles. The sum of the emission, for each ion stage recorded, at each observed kinetic energy, decreased with increasing angle from the target normal.

We report detailed studies of a threshold antenna frequency for the creation of an ion beam due to the formation of a stable electric double layer (DL) in an expanding, low pressure, argon helicon plasma. Mutually consistent measurements of ion beam energy and density obtained with a retarding field energy analyzer and laser-induced-fluorescence indicate that a stable ion beam of approximately 15 eV appears for antenna frequencies above 11.5 MHz. At lower antenna frequencies, for which the rf coupling to the plasma improves, large electrostatic instabilities appear downstream of the expansion region and a well-formed ion beam is not observed. Further studies of the low frequency (∼17.5 kHz) electrostatic fluctuations suggest that they arise from beam-driven, ion acoustic instabilities. We also observe a sharp increase in the upstream density at the same threshold antenna frequency, confirming a theoretical model that predicts an increase in the upstream density due to enhanced ionization resulting from electrons accelerated upstream by the DL.

A transonic argon plasma is studied in an integral simulation where both the plasma creation and expansion are incorporated in the same model. This integral approach allows for simulation of expanding plasmas where the Mach number is not known a priori. Results of this integral simulation are validated with semi-analytical models. Inside the creation region the results for the electron temperature, the heavy particle temperature and the electron density are compared with a global model of the creation region. In the expansion region, the simulation results of the compressible flow field are compared with predictions for the shock position. Both the results inside the creation region as well as in the expansion region are in good agreement with the semi-analytical models.

Using a time-resolved Langmuir probe the temporal evolution of the bulk plasma parameters in a high-power impulse magnetron sputtering (HiPIMS) discharge was investigated for a number of different discharge conditions. The magnetron was operated in argon between 0.5 and 1.6 Pa with a titanium target and with peak target power densities up to 1000 W cm⁻². The pulse width and repetition rate were held constant at 100 µs and 100 Hz, respectively. Using an OML analysis as well as a Druyvesteyn formulation, the electron densities, effective temperatures and energy distribution functions were obtained throughout the pulse period (0–9 ms), including a detailed study of the first 10 µs, which was achieved with a temporal resolution better than 0.5 µs. In the initial phase of the voltage pulse (t ∼ 1–4 µs), three distinct groups of electrons (indistinguishable from Maxwellian electrons) were observed, namely 'super-thermal', 'hot' and 'cold' populations with effective temperatures of 70–100 eV, 5–7 eV and 0.8–1 eV, respectively. After 4 µs these groups become energetically indistinguishable from each other to form a single distribution with an electron temperature that decays from about 5 to 3 eV during the rest of the pulse on-time. The presence of the 'super-thermal' electron group pushes the probe floating potential to a very negative value (significantly deeper than −95 V) during the initial period of the pulse. In the off-time, the electron density decays with two-fold characteristic times, revealing initially short-term (30–40 µs) and ultimately long-term (3–4 ms) decay rates. These long decay times lead to a relative high density remnant plasma (2 × 10⁹ cm⁻³) at the end of the off-time, which serves to seed the next voltage pulse. The electron temperature and plasma potential also exhibit two-fold decay in the off-time, but with typically somewhat faster decays, particularly for the long-term decay (100–500 µs) up to the end of the off-time. The time evolution of the plasma potential shows that for a considerable fraction of the on-time the plasma potential remains negative (down to −12 V) only becoming positive after t ∼ 60 µs which corresponds to a time of maximum plasma density (typical values of 2 × 10¹² cm⁻³). The generation of super-thermal electrons in the initial phase of the discharge is argued through the development of a simple magnetized-electron bounce model of the expanding sheath.

An RF-assisted closed-field dual magnetron sputtering system was developed to characterize the plasma and the ionization fraction of sputtered material to provide a suitable system for depositing optical thin films on large-area substrates at low temperatures (<130 °C). The 'prototype' system consists of dual 76 mm dc magnetrons operated at both balanced and unbalanced (closed-field) configurations with an RF coil to initiate a secondary plasma. The RF plasma assistance enhanced the electron density to one order of magnitude higher, increased the deposition rate and effectively enhanced the ionization fraction of the sputtered flux to above 80% as measured from the quartz crystal microbalance combined with electrostatic filters. Based on the prototype system, a large-scale RF-assisted system using two 9 × 46 cm linear magnetron cathodes was also developed and evaluated. Both systems were also tested for reactive deposition of indium tin oxide on both small-scale and large-area polyethylene terephthalate substrates with the actual substrate surface temperature monitored to be <130 °C. From both the evaluation results of the plasma characterization and deposition performance, the prototype and the large linear magnetron system were found to be suitable for reactive thin film deposition of compound targets that can be extended to various types of optical coatings.

This paper studies the sheath in front of a plane probe immersed in an electropositive/electronegative plasma in the low ionization regime. We analyze the dependence of the model on the electric potential and the position of the probe. As will be shown, this relationship is a necessary condition to form the sheath and allows us to determine the thickness of the sheath. The above-mentioned relationship provides the mathematical structure of the boundary layer for the sheath in a two-scale formalism as it is well known. It also shows that the Bohm criterion unambiguously determines the potential at the sheath edge in electronegative plasmas even when the electronegativity of the plasma corresponds to values for which the electric potential has an oscillatory behavior, against the commonly accepted view.

In this work, electrical and optical studies of SF₆ and SF₆/O₂ plasmas generated in a hollow cathode reactive ion etching reactor were performed using the Langmuir probe and optical emission spectroscopy techniques, respectively. We carried out an investigation aimed at understanding the influence of radio-frequency power, gas pressure and O₂ gas mixing ratio on plasma parameters, namely electron temperature, electron density and electronegativity, and also atomic fluorine density. The results indicate an increase of up to one order of magnitude in electron density and atomic fluorine in the overall gas volume when compared with a conventional reactive ion etching plasma generated under the same operation conditions.

A method is presented that enables measurement of plasma potential waveforms. The method consists of measurement by an uncompensated probe and analysis of the measured waveforms by means of a model of the sheath around the probe. The method was successfully tested in a nitrogen capacitively coupled discharge. The strong influence of the sheath around the probe on the probe voltage and the correlation between the probe current waveform and the discharge current are shown. At low pressures, the plasma potential waveform contains a large amount of higher harmonic frequencies whose amplitudes are comparable to the amplitude of the fundamental frequency.

A decrease in electron density and a strong increase of electron energy, which induce the enhancement of excitation rates, have been observed in CH₄–CO₂ plasmas when the inlet methane concentration is high enough and the input microwave power sufficiently low. Together with the decrease in the electron density with plasma duration, they are characteristic of dust formation in these plasmas. In these conditions, the formation of hydrocarbon radicals which are well known precursors of soot and the formation of first stable aromatics are reported, as observed by molecular beam mass spectrometry. Modelling of the chemistry in the plasma is carried out, which can also predict the formation of low concentrations of polyaromatic hydrocarbons. These species could be involved in the homogeneous nucleation process of carbon. As a function of the plasma duration, various carbon nanostructures are observed in the particles collected downstream of the plasma. For short durations, nanodiamond grains are formed with the size range 15–100 nm. They are composed of diamond nanocrystals of about 2–10 nm in size; these values are generally observed for all diamond nanocrystals formed in extraterrestrial and terrestrial conditions. For longer plasma durations, sp²-hybridized carbons are obtained. Their structure varies from soot to more ordered graphitic carbons nearly similar to 'onions' and structures similar to those observed in tokamaks. The control of the size and the microstructure of the nanodiamond grains are especially important as this could open possibilities for applications in a wide range of fields.

An experimental study of a direct-current, surface arc discharge in a Mach 2 cold supersonic airflow is presented. The surface arc discharge is generated with cylindrical tungsten electrodes flush-mounted on a boron-nitride ceramic plate embedded in the lower wall of the supersonic test section. In the presence of airflow, gas breakdown voltage increases from 1.5 kV in stationary air to 2 kV due to particle number density augmentation in the flow. The surface arc discharge transforms from a continuous mode in stationary air to a pulsed-repetitive mode in the flow. The mean time interval between discharge pulses is about 4.3 ms. For a single pulse, arc discharge occupies only about 60 µs. The discharge photos taken by a high-speed CCD camera (framing rate 1125 Hz) validate this pulsed-repetitive process and indicate that the plasma channel of the surface arc discharge is blown downstream by the supersonic flow. As the length of the plasma channel increases, the discharge voltage also increases. When the channel length reaches a critical value (∼25 mm), the dc power supply (3 kV–4 kW) cannot sustain the discharge voltage (∼3 kV) and the Joule heating energy cannot balance the dissipation of constrained convection, and hence the discharge quenches immediately. Current and voltage measurements demonstrate that the discharge process in a single pulse can be separated into three distinct phases: strong-pulsed breakdown process, steady discharge process and discharge attenuation process. Finally, the underlying mechanism of the dynamic process of surface arc discharge in supersonic flow is discussed. This paper provides more insights into the mechanism of supersonic flow control (in particular, shock waves) by a surface arc discharge.

A transverse flow dc glow discharge in Ar/CH₃I mixtures was modeled using a plug-flow approximation and plasma evolution was computed within a frame of reference moving with the gas. Reduced electric field as a function of transportation length was determined using the condition of a constant power density in the discharge. Cathode fall (CF) was found by comparing modeled and experimental discharge voltages. It was shown that conditions in the plasma column are nearly optimal for iodine atoms production with the lowest required energy and larger energies observed in the experiment are due to energy loss in the CF. In the experiment, concentrations of iodine molecules were measured downstream of the discharge using laser-induced fluorescence. Iodine flow at the exit of the discharge generator consisted of 80–90% of iodine atoms and only of 10–20% of iodine molecules. A good agreement between modeled and measured concentrations was found.

A detailed global model of atmospheric-pressure He + H₂O plasmas is presented in this paper. The model incorporates 46 species and 577 reactions. Based on simulation results obtained with this comprehensive model, the main species and reactions are identified, and simplified models capable of capturing the main physicochemical processes in He + H₂O discharges are suggested. The accuracy of the simplified models is quantified and assessed for changes in water concentration, input power and electrode configuration. Simplified models can reduce the number of reactions by a factor of ∼10 while providing results that are within a factor of two of the detailed model. The simulation results indicate that Penning processes are the main ionization mechanism in this kind of discharge (1–3000 ppm of water), and water clusters of growing size are found to be the dominant charged species when the water concentration is above ∼100 ppm. Simulation results also predict a growing electronegative character of the discharge with increasing water concentration. The use of He + H₂O discharges for the generation of reactive oxygen species of interest in biomedical applications and the green production of hydrogen peroxide are also discussed. Although it would be unrealistic to draw conclusions regarding the efficacy of these processes from a zero-dimensional global model, the results indicate the potential suitability of He + H₂O plasmas for these two applications.

Multiple steady-state solutions existing in the theory of dc glow discharges are computed for the first time. The simulations are performed in 2D in the framework of the simplest self-consistent model, which accounts for a single ion species and employs the drift–diffusion approximation. Solutions describing up to nine different modes were found in the case where losses of the ions and the electrons due to diffusion to the wall were neglected. One mode is 1D, exists at all values of the discharge current, and represents in essence the well-known solution of von Engel and Steenbeck. The other eight modes are axially symmetric, exist in limited ranges of the discharge current, and are associated with different patterns of current spots on the cathode. The mode with a spot at the centre of the cathode exhibits a well pronounced effect of normal current density. Account of diffusion losses affects the solutions dramatically: the number of solutions is reduced, a mode appears that exists at all discharge currents and comprises the Townsend, subnormal, normal and abnormal discharges. The solutions that exist in limited current ranges describe patterns, and these patterns seem to represent axially symmetric analogues of the 3D patterns observed in dc glow microdischarges in xenon.

It is known that the presence of residual gases can significantly suppress the sputter erosion rate of surfaces. Concerns about the validity of witness plate measurements are raised when this happens in ground-based tests of plasma thrusters. Here we quantify the influence residual gases impose upon measurements made with witness samples placed within the energetic ion plume of a Hall-effect plasma thruster. Specifically, sputter rates of thin films composed of iron by xenon ions are presented as functions of ion energy and partial pressure of molecular oxygen and nitrogen. When gas contaminates are present, the sputter rate is determined as a function of the poisoning ratio, R_p, defined as the rate of arrival of contamination sticking to a surface divided by the removal rate of sputtered material due to ion bombardment. The iron sputter rate was measured versus poisoning ratio from 0.001 to 10. At low poisoning ratios, ion bombardment maintains a dynamically clean surface, sputtering is not suppressed and sputtering rates can be estimated from the literature or from experiments. At poisoning ratios of R_p > 0.1 however, the net sputtering rate can be attenuated. The sputter rate of iron in the presence of oxygen at R_p > 1 was found to be only 5.4% (0.054) and 11.5% (0.115) of the sputter rate at R_p < 0.01 by 300 eV and 600 eV xenon ions, respectively. Although oxygen reduced the iron sputter rate, nitrogen was not found to affect the sputter rate of iron.