Table of contents

Interest has grown over the past few years in applying atmospheric pressure plasmas to plasma processing for the benefits this can offer to existing and potential new processes, because they do not require expensive vacuum systems and batch processing. There have been considerable efforts to efficiently generate large volumes of homogeneous atmospheric pressure non-thermal plasmas to develop environmentally friendly alternatives for surface treatment, thin film coating, sterilization, decontamination, etc.

Many interesting questions have arisen that are related to both fundamental and applied research in this field. Many concern the generation of a large volume discharge which remains stable and uniform at atmospheric pressure. At this pressure, depending on the experimental conditions, either streamer or Townsend breakdown may occur. They respectively lead to micro-discharges or to one large radius discharge, Townsend or glow. However, the complexity arises from the formation of large radius streamers due to avalanche coupling and from the constriction of the glow discharge due to too low a current. Another difficulty is to visually distinguish many micro-discharges from one large radius discharge. Other questions relate to key chemical reactions in the plasma and at the surface. Experimental characterization and modelling also need to be developed to answer these questions.

This cluster collects up-to-date research results related to the understanding of different discharges working at atmospheric pressure and the application to polymer surface activation and thin film coating. It presents different solutions for generating and sustaining diffuse discharges at atmospheric pressure. DC, low-frequency and radio-frequency excitations are considered in noble gases, nitrogen or air. Two specific methods developed to understand the transition from Townsend to streamer breakdown are also presented. They are based on the cross-correlation spectroscopy and an electrical model.

A remote atmospheric pressure discharge working with ambient air is used for the near room temperature treatment of polymer foils and textiles of varying thickness. The envisaged plasma effect is an increase in the surface energy of the treated material, leading, e.g., to a better wettability or adhesion. Changes in wettability are examined by measuring the contact angle or the liquid absorptive capacity. Two regimes of the remote atmospheric pressure discharge are investigated: the glow regime and the streamer regime. These regimes differ mainly in power density and in the details of the electrode design. The results show that this kind of discharge makes up a convenient non-thermal plasma source to be integrated into a treatment installation working at atmospheric pressure.

Micro-structured electrode (MSE) arrays allow the generation of large-area uniform glow discharges over a wide pressure range up to atmospheric pressure. The electrode widths, thicknesses and distances in the micrometre range are realized by means of modern micro-machining and galvanic techniques. The electrode distance, the gap width d, is small enough to generate sufficiently high electric field strengths to ignite gas discharges by applying only moderate radio frequency (RF, 13.56 MHz) voltages (80–390 V in Ne, He, Ar, N₂ and air). The non-thermal plasma system is characterized by a special probe measuring the electric parameters. We tested MSE arrays with d = 70, 25 and 15 µm. The MSE driven plasmas show a different behaviour from conventional RF discharge plasmas. Due to the very small electrode gap width we can describe the behaviour of the charged particles in the RF field of our system with the dc Townsend breakdown theory, depending on the pressure range and gas. With decreasing pressure, the gas discharges, especially in Ne and He, are increasingly dominated by field electron emission. With the MSE arrays as plasma sources several applications were developed and successfully tested, e.g. decomposition of waste gases and sterilization of food packaging materials at atmospheric pressure.

The techniques of spatially resolved cross-correlation spectroscopy (CCS) and current pulse oscillography were used to carry out systematic investigations of the barrier discharge (BD) in the binary gas mixtures N₂/O₂ at atmospheric pressure. At very low oxygen concentrations (<500 ppm), the BD was observed in a so-called diffuse mode (also referred to as atmospheric pressure glow discharge, glow silent discharge or homogeneous BD). In the case of the BD filamentary mode, the spatio-temporal distributions of the BD radiation intensities were recorded for the spectral bands of the 0–0 transitions of the second positive (λ = 337 nm) and first negative system of molecular nitrogen (λ = 391 nm). In the case of the diffuse mode, the spectral bands λ = 337 nm, λ = 260 nm (0–3 transition of the γ-system of NO) and λ = 557 nm (radiation of ON₂ excimer) were used for this purpose. The velocities of the cathode-directed ionizing waves as well as the effective lifetimes of the excited states and were evaluated from the CCS data. Special attention was devoted to the investigation of the transition between the filamentary and diffuse modes of the BD, this transition being caused by the variation of oxygen content within the range 500–1000 ppm.

This work is a contribution to the understanding of the mechanisms controlling the transition from a Townsend to a filamentary dielectric barrier discharge when the power increases. The approach consists in developing an electrical model of the discharge and the power supply to study the interaction between these two elements. The main components of the discharge model are (i) two Zener diodes whose characteristics depend on the power to take into account the effect of the gas density variation induced by the temperature variation and (ii) a RC circuit describing the memory effect from one discharge to the following one which is due to the metastables and the electrons trapped on the surface of the solid dielectrics. The calculated and measured currents are very similar over all the range of amplitude and frequency allowing to get an atmospheric pressure Townsend discharge. The model also describes the transition to filamentary discharge observed when the excitation frequency increases too much showing that it is due to a very fast variation of the load connected to the power supply. From this understanding, a solution is deduced to increase the maximum power dissipated in the discharge which consists in decreasing the solid dielectric capacitance.

We present experimental results of 'atmospheric pressure glow discharges' in two different electrode configurations, plane–plane and cylinder–plane. Although many of the phenomena we report here have been observed in different noble gases, we restrict this presentation primarily to the case of helium. Discharge diagnostics have been carried out using ultra-high speed imaging, and synchronous detection of light emission and current–voltage measurements, the former using a photomultiplier. The discharge physics is discussed with reference to recent work reported by the present authors and others.

A transparent plasma discharge reactor using air was used to investigate the transition from a filamentary dielectric barrier discharge (FDBD) operation regime into the diffuse barrier discharge regime. Recent results of other researchers indicate that the stability of diffuse barrier discharges in nitrogen may be attributed to the lack of a fully formed cathode fall layer when the discharge operates in a regime between the Townsend discharge and a normal glow discharge. We have demonstrated that a diffuse barrier discharge in air exhibits an increased accumulation of electric charge on the electrode's dielectric plates as compared with the FDBD. This may provide a means of stabilizing the discharge in a Townsend-to-glow discharge transition. Unlike operation in nitrogen, a streamer mechanism is involved in the formation of a uniform air plasma, though in a different manner than is associated with the FDBD. Numerous diffuse streamer clusters were observed on pre-charged dielectric plates at the breakdown voltage. Our conclusion is that the macroscopically uniform atmospheric-pressure DBD in air is obtained by the numerous radially expanding streamers that are temporally overlapping.

In this paper, we present data on the physics and phenomenology of plasma reactors based on the One Atmosphere Uniform Glow Discharge Plasma (OAUGDP™) that are useful in optimizing the conditions for plasma formation, uniformity and surface treatment applications. It is shown that the real (as opposed to reactive) power delivered to a reactor is divided between dielectric heating of the insulating material and power delivered to the plasma available for ionization and active species production. A relationship is given for the dielectric heating power input as a function of the frequency and voltage at which the OAUGDP™ discharge is operated.

The aim of this work is to develop barrier coatings by using an atmospheric pressure dielectric barrier discharge (DBD). For this purpose the plasma polymers obtained from a hybrid organic–inorganic precursor compound as well as from a purely organic compound were compared. The physical properties of the coatings were determined by scanning electron microscopy and scanning probe microscopy analysis as well as by profilometry and by determining their oxygen transmission rates. The chemical structure was revealed by ATR-FTIR analysis and ¹³C- and ²⁹Si-CP/MAS solid-state nuclear magnetic resonance measurements. It is shown that atmospheric pressure plasma polymerization of hybrid polymer precursors leads to coatings with excellent barrier properties due to the unique synergistic effect of the organic and the inorganic network structures that are formed in the plasma.

A review is presented of the studies in the former Soviet Union and in the USA of the mutual interactions of plasmas and high speed flows and shocks. There are reports from as early as the 1980s of large changes in the standoff distance ahead of a blunt body in ballistic tunnels, significantly reduced drag and modifications of travelling shocks in bounded weakly ionized gases. Energy addition to the flow results in an increase in the local sound speed that leads to expected modifications of the flow and changes to the pressure distribution around a vehicle due to the decrease in local Mach number. The critical question was, did a plasma provide a significant energy multiplier for the system? There have been a large number of experimental studies on the influence of a weakly ionized plasma on relatively low Mach number shocks and inherently also on the influence of the shock on the plasma. This literature is reviewed and illustrated with representative examples. The convergence through more controlled experiments and improved modelling to a physics understanding of the effects being essentially due to heating is outlined. It is demonstrated that the heating in many cases is global; however, tailored experiments with positive columns, dielectric barrier discharges and focused microwave plasmas can produce very localized heating. The latter appears more attractive for energy efficiency in flow control. Tailored localized ionization and thermal effects are also of interest for high speed inlet shock control and for producing reliable ignition for short residence time combustors, and work in these areas is also reviewed.

The current density–electric field intensity relationship of pre-heated potassium tetrachlorozincate (KZC) crystals (undoped and doped with Mn²⁺ in different concentrations) has been measured along the polar a-axis. The dependence was studied at selected temperatures covering the three high-temperature phases of KZC in order to investigate the type of conduction mechanism dominating in each phase. The original Richardson–Schottky (R–S) equation shows disagreement between calculated and experimental values of the Richardson and optical dielectric constants. The modified R–S equation fits the data well and facilitated the calculation of the barrier height, electronic mobility and high-frequency dielectric constant of KZC. The calculated parameters are in good agreement with the corresponding experimental values. The results indicated that the bulk- and electrode-limited mechanisms contribute to conduction in different phases of KZC. The temperature dependence of the dc conductivity along the polar axis of undoped and Mn²⁺-doped KZC shows anomalous behaviour in the region of the phase transitions. The dc conductivity and the activation energy of conduction (w) changed significantly due to doping. As indicated by the extremely high w values, superionic conduction is the dominating mechanism in the high-temperature part of the incommensurate (IC) and normal phases of KZC. Suppression of the superionic conduction by Mn²⁺-doping is observed. The effect of discommensurate pinning by Mn²⁺ ions and the possibility of dislocation formation on the dc conduction in the IC phase is also discussed.

The dependence of the polarization of ferroelectric perovskite thin films on the film thickness is explicitly calculated by using a generalized Landau–Ginzburg–Devonshire thermodynamic theory. Both residual stress and stress relaxation in the films have a significant effect on the thickness-dependent polarization. Our calculations demonstrate that the coexistence or competition between the surface effect and the constraint effect results in significant size effects at a finite nanoscale. This is especially so for the compressively strained perovskite films, in which there are two characteristic thicknesses, i.e. h_c (a few nanometres), below which the ferroelectricity disappears, and h_m (∼10h_c), where the polarization exhibits a maximum. The theoretical results describe the available experimental data quite well.

The electromagnetic band gap properties through a two-dimensional plate constructed with centimetre-sized spherical foam balls surrounded by aluminium rings are studied in the microwave range of 0.7–18 GHz. We observe from both experiments and finite-difference time-domain simulations that multiple band gaps appear and correspond to different resonant modes localized in the rings. The principal stop band, the lowest mode, appears at low frequencies and is robust to any lattice structure, even disorder. However, the remaining auxiliary stop bands, caused by high-order resonances, are usually located at higher frequencies and are more sensitive to the configuration of the structure formed by the ring resonators. We note that the stop bands would be broadened and strongly attenuated if more layers of the same plates were packed together. In addition, even or odd modes can be excited under different orientations of the rings with respect to the polarization of the illuminating radiation.

A two-dimensional numerical code is developed to model the effect of grains boundaries (g.b.) on the capacitance–voltage (C–V) characteristics of polysilicon diodes. The test structures are lateral polysilicon P⁺N diodes where the thickness of the film t_f deposited by the low-pressure chemical vapour deposition method is 700 or 450 nm. The P⁺ and N dopings were performed by ion implantation using, respectively, boron in a dose of 2 × 10¹⁵ cm⁻² and phosphorus in two doses of 10¹⁴ cm⁻² for t_f = 700 nm and 5 × 10¹⁴ cm⁻² for t_f = 450 nm. Using scanning electron microscopy, the presence of one grain boundary (g.b.) perpendicular to the metallurgic junction and localized at 100 nm of the interface deposition has been observed. In this work, we particularly investigate the effect of this g.b. on the C–V characteristics. The measured C–V characteristics at 100 kHz and 1 MHz show that the frequency effect is more important in the case of the weakly doped film (t_f = 700 nm). A determination of the series resistance gives the profile doping concentration: abrupt (N_D = 5.5 × 10¹⁸ cm⁻³) for t_f = 700 nm and gradual (slope = 5 × 10²⁵ cm⁻⁴) for t_f = 450 nm.

Using the previous experimental parameters in the two-dimensional simulation, we show that the presence of the perpendicular g.b. can reduce by up to 25% the capacitance of the diode and decreases considerably the V_RP voltage that corresponds to the realizing–pinning of the electrostatic potential at the first parallel g.b. This effect is more important when the doping is gradual. The fit of the experimental curves gives, in the weak doping case (t_f = 700 nm), the position of the first parallel g.b. (L_G1 = 37.5 nm) and the density of the inter-granular trap states (N_T = 3.2 × 10¹² cm⁻²). On the other hand, when the doping is relatively strong (t_f = 450 nm), the fit shows that the C–V characteristic is dominated much more by the doping profile than by the position of the first grain boundary and the density of the inter-granular trap states.

Efficiency optimization of a stable and debris free plasma source at 13.5 nm, is at the forefront of current extreme ultraviolet lithographic (EUVL) research efforts. To date, 1–2.5% soft x-ray conversion efficiencies (CEs) within a 2% bandwidth (BW) around 13.5 nm and into 2π steradians have been attained experimentally for laser-produced plasmas containing Sn at power densities of 0.5–5 × 10¹¹ W cm⁻². In order to complement these experimental endeavours, we have undertaken to study the CE, for the given wavelength regime, in the optically thick limit. We have achieved this by coupling time-dependent and steady-state collisional-radiative (CR) equations to the output of the one-dimensional hydrodynamic code MED103 (MEDUSA), where a solid sphere of radius 50 µm was uniformly irradiated by a high intensity laser pulse with a Gaussian temporal profile. The ion populations obtained from these CR results were then used in an integrated spatio-temporal figure of merit (FOM) together with in-band weighted dipole oscillator strengths and transition energies. The maximum FOM, when divided by the laser energy, was found to occur in the range of peak power densities of 2–3 × 10¹¹ W cm⁻² for the steady-state and time-dependent models, respectively. The hydrodynamic variables of these peak power densities were then used in a radiative transfer calculation in which the many-celled spherical plasma was treated as a multi-component blackbody. It is found that CEs of 3.5–6% within the 2% BW per 2π steradians may be achieved. These results are of particular relevance to EUVL technologies where a minimum CE of 3% is required by industry.

We use 1 mm plane to plane configurations, with or without dielectric barriers, on metallic electrodes to investigate the ac electric precipitation of charged submicronic particles of controlled electrical mobility injected in a laminar flow. The collection efficiency is calculated from concentration measurements for different voltages. Below the air ionization threshold, experimental and theoretical collection efficiencies are consistent. Above the air ionization threshold, the current and collection efficiency increase with frequency and voltage related to the number of microdischarges per second and per unit surface area. This study intends to study the influence of surface polarization due to the development of microdischarges on the collection of charged particles. Comparison between the theory of collection in a homogeneous ac field without microdischarge, on the one hand, and measured collection efficiency with microdischarges, on the other hand, proves that surface polarization by microdischarges does not improve the collection efficiency, but enables either an increase of charge by ion repulsion from the surface, whatever the frequency is, or an improved mixing at 60 kHz.

An atmospheric pressure glow discharge (APGD) was used for surface modification of polyethylene (PE) and polypropylene (PP). The discharge was generated between two planar metal electrodes, with the top electrode covered by a glass and the bottom electrode covered by the treated polymer sample. The discharge burned in pure nitrogen or in nitrogen–hydrogen or nitrogen–ammonia mixtures. The surface properties of both treated and untreated polymers were characterized by scanning electron microscopy, atomic force microscopy, surface free energy measurements and x-ray photoelectron spectroscopy. The influence of treatment time and power input to the discharge on the surface properties of the polymers was studied. The ageing of the treated samples was investigated as well. The surface of polymers treated in an APGD was homogeneous and it had less roughness in comparison with polymer surfaces treated in a filamentary discharge. The surface free energy of treated PE obtained under optimum conditions was 54 mJ m⁻² and the corresponding contact angle of water was 40°; the surface free energy of treated PP obtained under optimum conditions was 53 mJ m⁻² and the contact angle of water 42°. The maximum decrease in the surface free energy during the ageing was about 10%.

A simple method based on a conventional solid-state process is proposed for synthesis of manganese oxide nanowires. This method provides an opportunity for bunching the nanowires synthesized, since the nanowires form in one direction. Thus, aligned bunches (e.g. >100 µm long and about 50 µm in diameter) consisting of individual nanowires (diameters ranging between 50 and 200 nm) can be prepared simply. The phenomenon of simultaneous alignment of the nanowires in one direction to form ordered bunches is indeed an interesting development in the realm of nanotechnology. It can be considered as an alternative to conventional template-based methods.

Micrometre-scale Si pillars are formed by chemically enhanced laser ablation using nanosecond excimer laser irradiation of a Si single crystal in the presence of SF₆. We demonstrate the importance of precursor holes in determining the positioning of the pillars and show that we can control the initiation of precursor holes by ruling a grating into the Si substrate prior to irradiation. A rule defines an edge from which the laser light diffracts. Near-field amplification of the laser intensity enhances the formation of the precursor holes and aligns them parallel to the rule. The pillars can be thinned and eventually removed by wet chemical etching in aqueous KOH, resulting first in ordered arrays of extremely high aspect ratio pillars (e.g. tens of micrometres in length, with ∼ 10 nm tips) and then macropores. The shape of the macropore is determined by crystallography and the anisotropy of the wet etchant.

Thermally deposited 200 nm polycrystalline films of lithium fluoride (LiF) grown on glass substrates were irradiated with 150 MeV Ag ions at various fluences between 1 × 10¹¹ and 2 × 10¹³ ions cm⁻². The irradiation induced structural and optical modifications were studied using glancing angle x-ray diffraction (GAXRD), optical absorption and photoluminescence (PL) spectroscopy. The GAXRD results show that the films are polycrystalline and the average grain size (estimated from the widths of the GAXRD peak using the Scherrer formula) decreases systematically from 46.3 nm for the pristine sample to 18.3 nm for the sample irradiated at a fluence of 3 × 10¹² ions cm⁻². Thereafter, it remains constant. This reduction is attributed to strain induced fragmentation of grains. The optical absorption studies show dominant absorption bands of F₃ (385 nm) and F₂ (445 nm) colour centres. It is observed that the concentration of the colour centres increases with ion fluence and gets saturated at higher fluences. This can be correlated with GAXRD results in the sense that as the density of grain boundaries increases the concentration of colour centres also increases. The variation with fluence in PL intensities of the F₂ and colour centres is studied. The intensity of both bands (F₂ and ) increases up to a fluence of 1 × 10¹² ions cm⁻², followed by an exponential decrease, which is due to the increase in the non-radiative transition rate in the presence of defect-rich material.

A sodium bismuth titanate (Na_0.5Bi_0.5)TiO₃ (NBT) thin film of perovskite structure, synthesized by radio-frequency magnetron sputtering, exhibited a remanent polarization of 11.9 µC cm⁻² and a coercive field of 37.9 kV cm⁻¹ at room temperature. It however showed a rather high leakage current density of ∼6 × 10⁻⁵ A cm⁻² at an applied electric field of 100 kV cm⁻¹. There occurs a change in the controlling mechanism of the electrical behaviours of the NBT thin film from grain interiors to grain boundaries with increasing temperature. The ac conductivity obeys the Jonscher relation. The activation energies for dc conductivity and hopping frequency of the charge carriers are calculated to be 0.92 eV and 1.00 eV, respectively, suggesting oxygen vacancies are the most likely charge carriers at high temperatures. Hopping of oxygen vacancies trapped at the grain boundaries and excitation of polarons in the grain interior are responsible for the relatively high dielectric loss and high dc conductivity. The contribution of hopping charge carriers to the dielectric response is demonstrated by the frequency dispersion observed for the relative permittivity in the low frequency region.

Cellular polypropylene (PP) films were treated with sulfur hexafluoride (SF₆) gas in order to study the SF₆ penetration behaviour and optimize the electric charging conditions. There were differences in the penetration of SF₆ for different cellular PP materials, depending on the microscopic properties, which manifest themselves in the voided structure as well as in the mechanical stiffnesses of the cellular films. The penetration of SF₆ after long-term pressure treatment is confirmed in strongly inflated cellular PP films with a low mechanical stiffness of about 1 MPa. No SF₆ penetration occurs for slightly inflated cellular PP films with smaller void sizes and higher mechanical stiffnesses of around 5.8 MPa. The observed thickness variations, the higher charging fields during corona charging because of SF₆ penetration and the SF₆ environment, as well as the resulting electromechanical properties are discussed.

It is well known that the efficiency of material removal mechanisms has a crucial influence on the performance and quality of the laser cutting process. However, they are very difficult to study since the physical processes and parameters which govern them are quite complicated to observe and measure experimentally. For this reason, the development of theoretical models to analyse the material removal mechanisms is very important for understanding the characteristics and influence of these processes.

In this paper, a theoretical model of the pulsed laser fusion cutting of ceramics is presented. The material removal mechanisms from the cutting front are modelled under the assumption that the ceramic material may be, simultaneously, melted and evaporated by the laser radiation. Therefore, three ejection mechanisms are investigated together: ejection of molten material by the assist gas, evaporation of the liquid and ejection of molten material due to the recoil pressure generated by the evaporation from the cutting front.

The temporal evolution of the material removal mechanisms and the thickness of the molten layer are solved for several laser pulse modes. Theoretical results are compared with experimental observations to validate the conclusions regarding the influence of frequency and pulse length on the cutting process.