Search

Summary form only given. Amorphous silicon (a-Si) and silicon nitride (SiNx) etching process using atmospheric pressure plasma has been studied. As etchant, a small quantity of NF3 was mixed to nitrogen based DBD (Dielectric Barrier Discharge) plasma for the thin film etching. Test specimen was processed by passing through the downstream plasma zone under the specially designed plasma module with the speed of 50 mm/sec and with 3 mm of module-specimen spacing. The net treatment size of the plasma module (i.e. size of plasma jet nozzle) was 410 mm in width and 0.7 mm in length. The etch rate of the both film was increased with applied power and NF3 flow rate in the range of 0~3 lpm (liter per minute). The SiNx etch rate was dramatically increased by addition of oxygen at the elevated temperature with PR (Photo Resist) patterned specimen. At the optimized condition, the a-Si and SiNx etch rate was about 150 A/scan and about 200 A/scan at 50mm/sec scan speed, respectively. With PR patterned specimen the SiNx etch rate was increased about four times higher than the other case by addition of O2. In this experiment, the applied power range was 0.7~1.6 kW. And the etch rate results of the both layer did not show plateau in that power range.

Summary form only given. Mechanism and electrical characteristics of the discharge under 0.9% saline water have been studied. The discharge under water needs Ohmic heating to generate a void around the cylindrical electrode. The void around the electrode provides the discharge space at high voltage. Hence the electrical power dissipated in to two channels, heating the water to generate the void and generating the plasma in the void. The discharge power has been measured by a I-V Lissajous curves. After discharge in the void around the electrode, the void split into micro size bubbles and scattered in the water. The size of micro bubbles has a Gaussian distribution. However the average micro bubble size depends on the diameter of the electrode.

In this article, we examine the influence of contaminants on an electron cyclotron resonance discharge. For our discharge, the measured level of contaminant was highest just after startup, decreasing to a stable level after approximately 20 min.This is consistent with a limited source of gas trapped in the chamber walls. It is shown that the presence of these contaminants cause both the plasma and floating potentials vary by several volts. These potential variations are largest when the contamination is largest. Such variations were found in N2, CF4, and CHF3 discharges. The observed variation in the plasma potential should be of great concern in plasma‐aided manufacturing environments. The larger variations represent approximately 10% of the potential that one would typically apply between a device and the plasma. Such large changes in the potential might result in substantial changes in both etch rate and anisotropy. This can have particularly adverse effects on those devices having fine structures. Thus, one should monitor these contaminants and not process devices while the contaminant level varies.

SiO2 and Si etching in a CF4/O2/Ar plasma has been carried out in an electron cyclotron resonance etcher over a wide range of conditions. The etch rate has been compared with the ion energy flux to the wafer surface, JiEi, and the atomic fluorine density in the gas phase, nF. It is found that the etch rate can be divided into two regimes by a critical value of nF/(JiEi), the ratio of the atomic fluorine density to the ion energy flux. The critical value can be determined from a contour plot of the etch rate as a function of the ion energy flux and the atomic fluorine density. The critical value of nF/(JiEi) for Si is higher than that for SiO2. For nF/(JiEi) higher than the critical value, the SiO2 etch rate linearly increases with the ion energy flux, and the Si etch rate shows a nonlinear increase with the ion energy flux. For nF/(JiEi) lower than the critical value, both SiO2 and Si etch rates linearly increase with the atomic fluorine density.

Emissive probe measurements of plasma potentials outside the source aperture of an electron cyclotron resonance etching device have shown large radial electric fields in a nitrogen discharge. It is shown that the magnitude of the radial electric field is pressure dependent, decreasing for higher pressures. In addition to radial and axial profiles of the plasma potential, profiles of ion density, electron temperature, and floating potential were measured with Langmuir probes for several cases: Two magnetic‐field configurations, with and without an insulating liner in the source, and with and without a wafer stage. It is shown that the plasma potential with the liner present is higher when there is a resonance zone close to the aperture of the source, as opposed to when the resonance zone is farther inside the source.

Plasma flow velocity was determined experimentally from the time delay between fluctuations measured by two spatially separated Langmuir probes in both subsonic and supersonic, steady‐state, unmagnetized plasma flows.

The locations of excited hydroxyl radical [OH (A-X), in short OH*] generation are observed using the highspeed images in low-voltage-driven pin-to-electrolyte discharges. A comparison of the emission from the plasma and OH* reveals that the generation site of OH* is depending on the electrolyte polarity. It implies that the difference of OH* properties with polarity can be due to the OH* generation pathways on the electrolyte surface and in the plasma volume, corresponding to the humidity, the electron temperature, and the quenching rate.

The first direct measurement of a collisional Bohm presheath from plasma potential measurements is given. By measuring the presheath thickness in front of a grounded wafer stage, a determination of the collision mean free path for ions in an electron cyclotron resonance etching tool has been made. Presheaths were measured in N2 and CF4 plasma using an emissive probe. The presheath thickness in N2 was found to be linearly dependent on the mean free path. Measurements of CF4 plasmas, for which the collision cross sections are unknown, have shown results similar to those found for nitrogen. This result has enabled an extrapolation to be made of the effective cross section for collisions in plasmas created from CF4.

Many double-layer-relevant laboratory experiments have been carried out at the University of Wisconsin. Laboratory stair-step double layers, which resemble three or more weak double layers joined in series, have been produced without ionization. Double-layer floating potential fluctuations have been investigated and progress has been made in developing a novel technique for measuring electron energy distribution functions in low-density double layers (i.e. lambda /sub D/>>probe dimensions). A new inductive plasma source has been developed. With this source, magnetized double layers can be routinely produced. These magnetized double layers are often weak stair-step double layers that are oblique to the magnetic field. Laboratory data of emitting probe characteristics taken in tenuous plasmas have helped to quantify space-charge-enhanced plasma gradient induced error in double-layer electric field measurements made by satellite double probes. Also, magnetic sheaths have been experimentally studied and compared with theory. >

New features of the emissive probe characteristic curves in an inductive plasma source were observed in the E mode operation. More than the two inflection points in the derivative of the I–V characteristic curves are found when the harmonics of the input frequency (13.56 MHz) are present. The pairs of inflection points are symmetric about dc plasma potential.

Monitoring of plasma property becomes important to achieve the precise etching, deposition and other dry processes to fabricate nano scale semiconductor devices. To do this, the floating probe method for in-situ plasma monitoring tool has been developed under the consideration of less plasma interaction. Due to the electron energy distribution function (EEDF) the electron current flown through the electrical probe nonlinearly increases with the sheath potential between the probe and the plasma. Because of this nonlinearity, the probe current is a distorted sinusoidal current with many high frequency harmonics when a simple sinusoidal ac bias is applied to the floating probe. Because the amplitudes of these harmonics of the probe current strongly depend on the electron temperature and the density of plasma, the electron temperature and the plasma density can be obtained by means of the amplitude of current harmonics and its ratio. Due to the stray capacitance between the probe circuit and ground, the displacement current exists in the probe circuit. For more accurate measurement, the displacement current which consists in the probe current needs to be eliminated. In this study, the amplitude of sheath conduction current and the stray capacitance are obtained by means of the amplitude change due to the frequency dependency of displacement current. Comparing with the typical Langmuir probe diagnostic method, the floating probe with 500p/F of capacitance overestimats the electron temperature about 1eV higher and the plasma density about 150% higher when a 5kHz ac bias is applied to the probe. When the effect of displacement current is eliminated by means of the currents at 5kHz and 7kHz of bias frequency, by contrast, the floating probe with 500p/F of capacitance measured the electron temperature with 0.2eV of error and the plasma density with about 10% of error.

This paper introduces how to determine concentrations of ion species in a mixed gas plasma that are not linearly proportional to their neutral partial pressures. A particle balance model was developed to predict the relative ion concentrations in multiple-ion-species plasmas, considering their ionization rates and loss fluxes to the wall. Analysis is carried out especially with Ar/Xe and Ar/He multi-dipole plasmas in which the neutral gases are directly ionized by the mono-energetic primary electrons. The experimental data of ion concentrations were obtained using the ion acoustic wave measurement method of the concentration of two ion species. The comparison reveals that the ion concentration ratio is linearly proportional to the ratio of the ionization cross sections and the ion loss velocity between two gas species. Especially, the model prediction is improved with using the two-ion-species sheath model (recently reported by Baalrud and Hegna) for obtaining the ion loss velocity at the sheath boundary.

A global model was developed to investigate the densities of negative ions and the other species in a low-pressure inductively coupled hydrogen plasma with a bi-Maxwellian electron energy distribution. Compared to a Maxwellian plasma, bi-Maxwellian plasmas have higher populations of low-energy electrons and highly vibrationally excited hydrogen molecules that are generated efficiently by high-energy electrons. This leads to a higher reaction rate of the dissociative electron attachment responsible for negative ion production. The model indicated that the bi-Maxwellian electron energy distribution at low pressures is favorable for the creation of negative ions. In addition, the electron temperature, electron density, and negative ion density calculated using the model were compared with the experimental data. In the low-pressure regime, the model results of the bi-Maxwellian electron energy distributions agreed well quantitatively with the experimental measurements, unlike those of the assumed Maxwellian electron energy distributions that had discrepancies.

The analysis of the repetitive pulse discharge system for the plasma source ion implantation is investigated with both circuit simulation and experiment. In the circuit model, the ion and electron currents on a target are self-consistently varied with the applied voltage because the waveforms of repetitive pulse are affected by the internal properties of plasma, as well as the external circuit parameters. The circuit simulation reveals that not only the plasma properties, but also the circuit components, are important for pulse system to operate at high repetition-rate. The experiments are conducted with a plane electrode immersed in rf-driven argon plasmas. When negative high-voltage pulses are applied to the electrode, the current and voltage waveforms are measured and compared with the simulation results. Control parameters for high repetition-rate operation are discussed, based on the self-consistent circuit analysis of the pulse system.

Summary form only given. Analysis of the conduction current on the negatively biased target is important to develop the pulse switch system and to improve the material surface treatment using the plasma source ion implantation. In case of highly negative biased target immersed in the plasma, the current on the target is composed of the incident ion current and the electron emission current. The virtual area of collecting current on the target is proportional to the sheath area formed over the target electrode which is expanded considerably compared with the size of electrode with increasing the high voltage. The sheath developed near the edge of the target should be considered in the analysis of the target ion current. The spatial distribution of plasma in front of the target also affects the sheath formation because the density distribution is not uniform in real situation which may shrink the sheath expansion. In this study, the effects of sheath geometry and plasma distribution with the secondary electron emission are investigated. Experiments were carried out with the planar stainless steel and aluminum targets having a diameter 100 mm, negatively bias potential ranging in 2 kV-11 kV, and various plasma gases of Ar and He, respectively. Also the target is enshrouded by the quartz tube of 100 mm long which may reduce the effect of the sheath formation at the edge of the target. Spatial plasma density was measured by Langmuir probe and compared with the values obtained from the target current which were analyzed with the current model considered with the Bohm current, the edge effect of sheath formation on the target, and spatial plasma distribution with secondary electron emission coefficient. Results will be discussed

Summary form only given. In PSII, the conducting target is immersed in the plasma and the dose of implanted ions is controlled by the plasma density at the sheath edge. When evaluating the ion current or sheath formation, the stationary-bulk plasma density has been adapted. However the dynamic sheath is expanded with the applied pulse voltage on the target and the field in the sheath is enough to break down the gas. The emitted secondary electrons also employs on the perturbing the bulk plasma. Thus the sheath formation near the pulsed target should be influenced by the temporal plasma property with applying the voltage. The measurements are carried out in the target plasmas which are generated by the pulsed bias only and by the pulsed bias and the external RF power. Also the temporal property is obtained for the plasma generated by the pulsed RF plasma. The temporal plasma properties near the target and in the bulk are monitored by a RF compensated Langmuir probe. The current-voltage probe data is analyzed with the newly developed algorithm based on the biorthogonal wavelet transform. The properties, EEDF, electron temperature and density, are obtained with various pulse conditions and gases. Effects of the temporal and stationary plasmas on the Child-Langmuir sheath formation in front of the target will be discussed

To investigate the plasma recovery during the fall time of a high voltage pulse, a numerical model is developed by applying fluid equations for ions while assuming thermal equilibrium for electrons. In the model, effects of the circuit impedance of pulse system are taken into account. The numerical analysis reveals that the plasma recovery is affected by both the properties of plasma and internal impedance of pulse system. The experiments are conducted with a plane electrode immersed in rf-driven argon plasmas. When negative, high voltage pulses are applied to the electrode, the current and voltage wave-forms are measured and compared with the simulation results. Effects of internal circuit impedance as well as plasma properties on plasma recovery during pulse fall time are discussed based on the experimental and numerical results.

Summary form only given. The virtual boundary layer between the atmospheric discharge plasma and the dielectric barrier electrode was observed in atmospheric dielectric barrier discharge plasma. With considering the capacitance formed in the boundary layer, the accuracy of analyzing the dissipated power from the charge-voltage Lissajous diagram can be improved. Observations show that the boundary layer has the similar property of the sheath. The thickness of the boundary layer on the electrodes is proportional to the discharge density in the Townsend-mode discharge and to the applied voltage in the diffusive streamer-mode discharge, respectively. In the glow like discharge using the helium plasma, the thickness is proportional to the 2/3 power of applied voltage over the square root of the discharge density, which is very close to the collisional sheath property. The alpha-gamma transition effect on the boundary layer formed at the conducting electrodes is also investigated. Experiments were carried out with the parallel cell reactor having at least one of dielectric electrodes. The reactor capacitance was varied with changing the gap distance. The system is closed so that the gas was fed between the electrodes and exhausted the other side. The gas flow was controlled to minimize the perturbation of the discharge. Operating frequency is lower than a few hundred kilohertz. The property of the boundary plasma on the electrodes will be discussed

Electron energy distribution function (EEDF) from the Langmuir probe (LP) data is important in the analysis of accurate plasma density and electron temperature. The determination of EEDF requires the 2nd derivative of LP data with respect to the probe voltage, which may enhance the noises in the probe data seriously. In order to filter the noise effectively, the wavelet transformation technique, especially the bi-orthogonal wavelet transform (BWT), was adapted to separate the noises from the 2nd derivative LP data. The accuracy of algorithm based on the BWT was evaluated with various noise levels for the Maxwellian distribution electrons. It provides the EEDF without losing the important information that has an advantage of minimizing the operator's choices to precede the analysis of LP data.

In the case of highly negative biased target immersed in the plasma, the current on the target is composed of the incident ion current and the emission electron current. The virtual area of collecting current on the target is proportional to the sheath area formed over the target electrode which is considerable increased with applied high bias. The spatially distributed plasma also affects on the sheath formation, resulting in the target current. Consideration of the dimensional sheath formation and spatial plasma distribution with the secondary electron emission coefficient is conducted for analyzing the target current. Experiments were carried out with the planar stainless steel and aluminum targets having a diameter 100mm, negatively bias ranging in 2kv~11kV. Sheath size and spatial plasma distributions are measured by electrical probes. The current model including the dimensional sheath, spatial plasma distribution and secondary electron emission coefficient will be presented.

A new floating harmonics method is developed to determine the electron energy distribution shape in the non-Maxwellian plasmas. The slope and curvature of the electron energy probability function (EEPF) at the floating potential can be obtained by measuring amplitude ratios among the first three sideband harmonics of the probe electron current. Together with the generalized EEPF formula, the EEPF shapes of the non-Maxwellian plasmas are able to be determined from the slope and curvature. Here, the new method is experimentally verified and its results are compared with EEPF measurements using the Langmuir probe (LP) method. The effective electron temperature and the EEPF shape parameter obtained by the method give good agreements with those of the LP measurements.

Summary form only given. Doped nanotube is a possible candidate of a nano diode for the nanoelectronic device. In order to dope the N or P type elements into a carbon nanotube (CNT), it is necessary to carry on the deformation of the tube such as opening the end of the tube or cutting the side of the tube. After filling the elements, those opening areas of the tube should be closed through the reconstruction and welding processes. Here is introduced the ion irradiation method to open and close the tubes and to fill the tubes with elements. An inductively coupled plasma (ICP) was used as the process source, operating the pressure 1 mtorr of Ar or N2 for opening and closing the nanotubes, and adding Cs into the plasma for the doping process. Such low pressure kept the collisionless plasma condition which may help to control the direction of the irradiated ion onto the tube. Plasma density increased linearly with increasing 1 to 2 kW of the input RF power, resulting that the dose rate of the irradiated ions was controlled. Target nanotubes, the multi-walled carbon nanotubes (MWNTs) were laid on a metal substrate in the ICP, which was biased with 0~200 V DC. The irradiated ions were accelerated with the sheath potential, defined as the potential difference of the space potential and the substrate potential. The space potential was controlled with an inserted metal grid which was grounded. The energy of irradiated ions was estimated precisely by measurement of a sheath potential using the emissive probe and the ion energy analyzer (IEA). Plasma density was measured by a Langmuir probe. The morphology of ion irradiated CNT was observed with the field emission scanning electron microscope (FE-SEM). The structural analysis of filled CNTs was carried out with the high resolution transmission electron microscope (HR-TEM). Then the change of crystalline structure due to the ion irradiation was characterized by a Raman spectroscopy. Preliminary results show that the length of MWNT decreases with increasing the energy of irradiated ions. At low energy of irradiated ions ( 150 eV) and high dose. More results will be presented

Summary form only given. In this paper, we configured the open type DBD (dielectric barrier discharge) system. Discharge character was remained as a stable feature and its properties were successfully investigated. Power absorption was measured as product of voltage and current as a function of frequency under open-air atmosphere. We observed the existence of an optimum frequency for power absorption. It is suggested that the optimum frequency occurs because the phase difference between voltage and current decreases as frequency increases. Helium addition to O/sub 2/-discharge under open-air condition shows OES intensity changes as He flow rate. This means that generation rate of radical may change according to the gas species although inert gas is added. Physical meaning of the area of current and voltage trajectory on 1-V plain is supposed to reactive part. To increase the power efficiency, we should reduce the area, i.e. the phase difference of voltage and current. Contact angle is measured on PR (photoresist) surface after some treatment conditions. Helium addition on the main O/sub 2/-discharge results in the increase of contact angle. It means that the He addition changes the radicals generated from oxygen and affects the surface of PR.

Summary form only given. Doped nanotube has a potential step of a few eV at the junction responsible for the nonlinear conductivity so that the nano diode can be manufactured. In order to dope the N or P elements into nanotubes, it is necessary to open the end of the tubes. Here the ion irradiation method to open the end of the tubes was introduced. Experiments were carried out with various ion energies and fluxes of the irradiated ions to the multi-walled carbon nanotubes (MWNTs). The irradiated ion energy and flux can be controlled by the sheath potential and the ion plasma density near the tubes respectively. Emissive probe with the inflection point in the limit of zero emitting current method was applied to measure the sheath potentials and Langmuir probe was to measure the plasma densities near the tubes. MWNTs were grown in PECVD (Plasma Enhanced Chemical Vapor Deposition) reactor and the ion irradiation was carried out in the inductively coupled N/sub 2/ plasma. The morphology of irradiated MWNTs was studied by SEM and TEM images. It was observed that the catalyst was removed and the length of MWNTs linearly decreases with the incident ion energy and ion flux. The preliminary results show that the length of MWNTs decreases linearly with a ratio of 3.5 nm/eV for increasing the irradiation ion energy from 100 eV to 200 eV. Above 200 eV, almost of MWNTs were eliminated with remaining the small spot. With ion energies of 25/spl sim/100 eV, the catalysts were effectively removed from the top of MWNTs. Furthermore effect of the mass and flux of irradiation metallic ions on the deformation of MWNTs is presented.

Summary form only given, as follows. Hyper-thermal neutral beam sources are developing for the next generation semiconductor manufacturing device. The collimator in the device contacts the plasma to extract ion beam and exchange the incident ion beam to neutral beam resulted in Auger effect on the collimator surface layer. Hence, the neutral beam energy is proportional to the voltage difference between the plasma potential and bias voltage on the collimator. However, the biased collimator affects on the plasma parameters, especially the plasma potential In this presentation, the effects of the collimator biasing on the plasma parameters have been investigated by using various diagnostics tools, Langmuir probe, capacitive probe, emissive probe and grid ion energy analyzer. Ar and He plasma were used for the research. In addition, the effects of the collimator's physical size will be discussed on the plasma parameters.

Summary form only given, as follows. In this study, collisional effects on the property of time dependent sheath are experimentally observed. The experiment is carried out for the very thin cylindrical target with various pulse bias and operating pressures. The rise time of pulse and the operating pressure vary. Sheath propagation is observed using biased Langmuir probe to determine the plasma parameters and the sheath evolution. The experimental observation to the time dependent sheath with various pressures will be presented.

Summary form only given, as follows. There are many demands on monitoring the plasma density without perturbing to the plasma of the processing chamber. Here we present the possibility of monitoring the plasma density without using any electrical and optical probing systems. The newly developed monitoring method is based on the Godyak plasma impedance model, assuming the electron temperature is constant with power and operating pressure, the plasma density can be obtained from the balance between input and the plasma powers. The plasma impedance was obtained by measuring voltage and current supplied to the antenna, and the phase difference between voltage and current. To evaluate the method the plasma density obtained by the method is compared to the plasma density taken by a Langmuir probe at the center of reactor.

Summary form only given, as follows. The characteristics of DBD plasma were studied using optical emission spectroscopy (OES) and electrical signals of discharge with various flow-rate of oxygen. The diameter of copper electrode was 50 mm, the thickness of dielectric barrier was 3 mm and the distance of electrode was 2 mm. The O/sub 2/ flow-rate varied from 0 to 50 LPM. The OES was measured intensity of the each elements generated in the plasma. The square pulses with 1/spl sim/4.5kHz were applied to DBD reactor.

Summary form only given, as follows. In a dielectric barrier discharge, the amount of charge on the dielectric material of the electrode plays an important role in sustaining the plasma. The charge depends on the capacitance of the reactor. The operating frequency may change the charge condition of the electrode. Consequently the plasma property should depend on the capacitance of the reactor and the operating frequency. In this study, the effects of reactor capacitance on the discharge were investigated by varying the operating frequency and waveform of the applied voltage. The experiments were carried out with a planar type reactor at atmospheric pressure.

Summary form only given, as follows. High-density plasmas generated in a magnetized inductively coupled plasma attract applications such as high rate etching and deposition processes for semiconductor fabrication. Thus, it is very important to know the characteristics of the magnetized inductively coupled plasma. In this study, the ion mass, collisional and magnetic effects on the generation of a magnetized inductively coupled plasma were considered. The plasma was generated at 1356 MHz RF power with the Nagoya type (m=1) antenna in compact cylindrical reactors. Permanent magnets and electromagnets were arranged around the chamber to produce the magnetic field along the axis of the reactor.

Summary form only given, as follows. The Langmuir probe (LP) is one of the major tools used in plasma diagnostics because of its good measurement capability and simple structure. The various parameters of plasma, such as ion, electron saturation current, plasma potential, temperature and floating potential can be obtained from LP data. However, analyzing plasma parameters from LP data, the noise and its non-stationary state of signal would prevent from obtaining the plasma parameters with accuracy. These problems often take place and make us spend a lot of time to analyze the LP data. To overcome these problems, the analysis algorithm using wavelet transforms, which are currently studied and recognized as good tools of signal processing was developed.

Summary form only given, as follows. In this study, the ion beam and neutral beam energy was measure by using neutral particle analyzer (NPA) with a electrostatic polarizer.

Summary form only given, as follows. E to H mode transition is a typical nonlinear problem in an inductive Plasma sources. The mode transition has been explained in terms of power balance. In this presentation, the nonlinear equivalent circuit modeling has been studied. The linear transformer modeling has been used for inductive plasma source. However, linear transformer model can not explain the mode transition. The nonlinear circuit model based on the assumption that induction field is coupled nonlinearly to the plasma. When nonlinear coupling effects included in the linear transformer model, this nonlinear harmonic oscillation problem can be expressed by Duffing equation. The advantage of Duffing equation is hysteresis as well mode transition can be explained.

Summary form only given, as follows. Hyperthermal neutrals with energy levels between 2 eV and 100 eV can be utilized for a diagnostic tool for the boundary region of fusion plasmas, for the space environment on Earth orbiting spacecraft, and for a surface modification and material's processing. DEGAS 2, Monte-Carlo neutral transport code, follows the trajectory of each neutral in the pancake-style ICP source, in which the reflective neutrals are generated. The code, the plasma density at the reflector, and power coupling optimize the height of the ICP source to maximize the neutral flux out of the ICP source. Using DEGAS 2 code we can calculate the flux and energy distribution of the hyperthermal beam at interesting positions. The measurement of particle energy distribution using NPA (neutral particle analyzer) will be discussed. Also, Langmuir probe measurement in the slim plasma source and the removal of a PR (photo resistor) on silicon wafer as the evidence of extracting hyperthermal neutral beam out of the plasma source will be presented.

Summary form only given, as follows. New features of the ozonated water generator have been studied. Conventional methods make use of a ozonator to generate ozone and a diffuser to dissolve ozone to water. However, the diffuser is no more necessary in the direct dissolving system. The direct dissolving system is featured by dielectric barrier discharge in the two layer flows, water and gases. In this system, ozone gas is generated by the microdischarge in the gas layer and dissolved in the water layer, simultaneously. The concentration of dissolved ozone is quantitatively measured by use of the idometric method.. When 1.2 L water is pretreated by 24 g of potassium iodide (KI), the measured ozone concentration was about 100 ppm for 15 min discharging time.

Summary form only given. The frequency characteristics of dielectric barrier discharge (DBD) have been studied for optimizing discharge efficiency. The experiment was performed with various experiment conditions to find the relation between the discharge frequency and power consumption. There is the optimum discharge frequency, which minimizes the consumption of the electrical power. Physically, the optimum discharge frequency means the inverse of the charge build-up time on the dielectric materials. Hence, the optimum discharge frequency depends on the capacitance of a reactor, which is a function of the geometry of the reactor and dielectric materials, and discharge conditions, which depends on the discharge voltage, and gases. From the experimental results, a semi empirical relation on the optimum frequency can be expressed as a function of the discharge gap, dielectric materials, sort of electrodes, and discharge gases.

Summary form only given. Helicon wave generation in weakly magnetized inductively coupled plasmas (ICPs) with a conducting boundary has been issued in this experiment. the characteristics of a helicon wave coupled with various antennas (one turn loop antenna, two turn loop antenna, and modified Nagoya. type antenna) and a conducting boundary was studied measurements were carried out in a weakly magnetized ICP (500 mm in dia and 600 mm in length). Plasma densities were 10/sup 10/-10/sup 11//cm/sup -3/ and magnetic field up to 80 gauss. A floating B dot probe, compensated to the capacitively coupled E field, was employed to measure the wave field in the plasma. It was observed that m=0 and n=1 mode and m=1 and n=2 mode helicon waves generated in the plasma using one turn loop and two turn loop antennas respectively. In case of using the modified Nagoya type antenna, m=1 and n=2 mode helicon wave was observed. The longitudinal wave number was proportional to the B/n, where n is plasma density. It was in fairly good agreement between the experimental results and the model considered the large aspect ratio (chamber radius over the chamber length). Furthermore, it was found that a transverse wave number of the m=1 mode. Helicon wave depended on the aspect ratio of the boundary plasma.

Summary form only given. Spatial and temporal variations of plasma density, temperature and potential are measured in a simulated Plasma Display Panel (PDP) cell by using a fast-scanning electric probe system. A PDP simulator was made 200 times bigger than the actual cell. Effect of collisionality on the probe analysis is discussed and changes of plasma parameters due to different gases (Ar, He, Xe) are shown. Thermalizations of electrons are analyzed by comparing the data of single probe with those of triple probe. Time dependent plasma properties are also measured at 5 points within one period (20 kHz) using a single probe with sub harmonic sweep frequency (200 Hz) of applied AC power supply and 100 kHz sampling rate.

Summary form only given. In plasma source ion implantaion (PSII), the target is biased by successive negative voltage pulses with an intrinsic finite rise time to implant ions from plasma to target surface, resulting in a time-dependent sheath around target. In this work, the Langmuir probe was used to study the sheath motion around a planar target with various pulse conditions and plasma conditions in PSII. It was observed that the time-dependent sheath consisted of two parts: the ion matrix sheath development and the dynamic sheath motion. A new model for the ion matrix sheath development was proposed to explain the experimental results which showed discrepancies compared with Lieberman's model (Stewart and Lieberman, 1991) and the computer simulation. The ion matrix sheath development was in proportion to square root of pulse rise rate over plasma density and also to square root of pulse rise time. The dynamic sheaths were faster than the ion acoustic speed after the pulse rise and eventually slowed down to approximately a third of ion acoustic speed.

Summary form only given, as follows. Radial variations of ion densities are measured by a scanning electric probe in a large electron cyclotron resonance plasma source with diameter 160 cm and height 70 cm. Measured density profiles are explained by an analytic fluid model based upon the classical transport theory assuming the plasma generation along the resonance layer. Dependance of density profiles on the position of resonance layer and magnitude of magnetic flux along the radial direction are analyzed. Improvement of uniformity by changing the resonance position and application to the development of larger plasma sources will be discussed.

Summary form only given, as follows. Analysis algorithm of an electrical Langmuir probe data has been developed. Most algorithms requires the user specified input values for selecting the ion saturation, electron saturation, and intermediate regions to find the ion, electron density, plasma potential, and plasma temperature, respectively. However, the input values may cause a lack of reproducibility to analyze the probe data. When a noise is included into the probe signal, it is almost impossible to obtain the reliable plasma parameters from the probe data. In this work, a new algorithm is developed with the wavelet transforms to filter out the nose signal from the probe data and to determine the plasma parameters. A biorthogonal wavelet transform applies to determine the initial plasma potential and Daubechies wavelet is to obtain the electron and ion saturation currents. Electron energy distribution function (EEDF) or electron energy probability function (EEPF) can obtain clearly from the processed raw data by a biorthogonal wavelet transform. This algorithm is not necessary to require the user's input values to analysis, thus it can be reduced an uncertainty from data analysis. With artificial data set including a random noise, the performance of this programs is evaluated and the result shows more than 95% accuracy.

Summary form only given, as follows. Temporal variation of electrons and ions are measured in two different pulsed corona discharge reactors, which are from the dissociation of NO/sub x/ and SO/sub x/ from the flue gases of combustion engines and power plants. Industrial scale reactor with height of 3 m and diameter of 0.5 m was made for dissociation of 500 liters of flue gas per hour. The reactor is made as a three-stage cylinder-cylinder type with a rugged inner cylindrical electrode. Another reactor is scaled by one of tenth of the industrial reactor. Electric probes have been installed in both the electrodes in order to measure the density of charged particle along with floating potential. After correcting the contribution from the displacement current, effects of system parameters on the variation of charged particles and dissociation efficiency are studied along with numerical simulation.

The location of hydroxyl radical (OH*) generation in aqua plasma is investigated with the images of vapor bubble discharge in a conductive electrolyte (saline). A comparison of the visible-and UV-ray data taken from the Na+* and radical OH* reveals that Na+* is generated in both volume and surface of bubble while most of OH* is observed at the vapor surface. This result implies that the steamer-type of aqua plasma was generated in the vapor bubble covered over a metal electrode tip, and the charged particles disperse along the surface of the vapor bubble.

The effect of a standing wave on plasma density is observed in an inductively coupled plasma (ICP) with a single turn antenna whose circumferential length is much shorter than the wavelength of the driving frequency. The experiment is performed in an argon plasma operated at 5 mTorr with 13.56 MHz power applied to the single turn antenna with diameter of 300 mm. Measured plasma density near the ground termination of the antenna is higher than near the power input, which degrades the uniformity in the ICP source. It is analysed with the standing wave effect of the current and voltage on the antenna, which is carried out with a transmission line model based on the transformer circuit model of the ICP. The wavelength and the distribution of the inductively and capacitively coupled powers along the antenna are obtained from the model with the measured impedance of the antenna. The expected density distribution agrees well with the measured one, revealing that the wavelength is shortened and the nonuniformity of the plasma density becomes severe with increase of the input power.

Summary form only given, as follows. Rf discharges are produced in a conventional inductively coupled plasma (ICP) device (aspect ratio 1/R/spl sim/7 and 4 turn antenna) in the presence of an external DC magnetic field (up to 120 gauss). A common feature of MEICP is that the RF power is coupled to the plasma across a dielectric window (similar to an ICP) at 13.56 MHz and DC magnetic fields results in enhanced heating and excitation of the plasma electrons by electron cyclotron resonance. Unlike a conventional ECR, MEICP is generated inductively at the low frequency, 13.56 MHz, and resonated at low magnetic fields. At this low magnetic field, the electrons are magnetized but the ions are very weakly magnetized. The helicon wave length is longer than the chamber length for this experimental conditions. Preliminary results (taken from Ar plasmas with 0.5-1 mTorr and 50-200 Watts) show ion plasma densities increase factor of 10 higher at B/spl sim/20 gauss (i.e., /spl omega//sub ce//spl sim/4 /spl omega//sub rf/) compared to those without magnetic or with larger magnetic field. Floating potentials are varied with magnetic field strength and along the axis.

SiO2 was etched on 300 mm wafers in a dual-frequency capacitive plasma reactor to study etch rate nonuniformity as a function of driving frequency and power. It is shown that the etch rate profile shape varies significantly with the driving frequency. It also is shown that for different driving frequencies, the behavior of etch rate profile shape with the power is quite different, namely: (i) for lower frequency (27 MHz), the shape almost does not change with the power; (ii) for higher frequency (100 MHz), the shape considerably varies with the power. These results clearly indicate that the main reason for the etch rate nonuniformity in high-frequency capacitive reactors is the plasma nonuniformity caused by electromagnetic (standing wave and skin) effects. Using a specially shaped top electrode rather than the traditional flat one is shown to considerably improve the etch rate uniformity.

The generation condition of sub-mm sized non-thermal plasmas (named aqua-plasmas) inside electrically conductive liquids (saline) is investigated with various electrical conductivities of the electrolyte. The breakdown condition of an RF-driven (380 kHz and less than 50 W) aqua-plasma discharge on a metal electrode is investigated using both experimental and numerical simulation methods. Since breakdown occurs in the vapor covering the powered electrode surface, the boiling process and the influence of fluidic dynamics on the vapor coverage oscillation are considered in detail. The vapor coverage is formed on the metal electrode surface via three phases: individual bubble generation, merging of bubbles and maintaining the bubbles during the aqua-plasma discharge inside the full vapor coverage. The surface temperature is evaluated from a numerical simulation based on the proposed one-dimensional aqua-plasma equivalent circuit electro-fluid power balance, and the bubble coverage can be sustained due to boiling of the film on the heated electrode by a current flow of electrolyte ions, especially Na+ in this case. The periodic discharge of the aqua-plasma inside the shrunken vapor coverage is investigated by a model that considers the electrolyte viscosity and surface charge model. The vapor coverage periodically shrinks under the heat balance between the gain, mainly plasma heating and heat flux from the heated tip, and the loss to the saline, so it is a feature that the aqua-plasma is discharged during the period of shrinkage. Moreover, the tip, heated up to meet the temperature condition of film boiling ~300 °C, reduces the breakdown field due to the reduced work function of the electrode material.

Iodine and cesium atoms were encapsulated in single-walled carbon nanohorns (SWCNHs). Atom encapsulation was carried out with sequential plasma aided procedures which consisted of opening SWCNH tips with an oxygen plasma, atom insertion in an iodine-mixed or cesium-mixed argon plasma, and closing the open tip in an argon plasma. Results reveal that oxidation plays a role in the tip opening procedure, and capillary forces are the driving force for the permeation of the atoms through the open tip of the SWCNHs. The open tip of the atom inserted SWCNH can be closed under the ion irradiation. It demonstrated the fabrication process of encapsulating atoms in SWCNH by using the sequential plasma assisted processes.

Dynamic sheath expansion in front of a target biased with a negative, high-voltage pulse is investigated in a non-uniform plasma, taking into account the influence of ion drift, which is inevitable in diffusive plasmas with a non-uniform density profile. Temporal evolutions of a sheath edge and a rarefaction wave measured in a low-pressure argon plasma diffusing towards the target agree well with numerical predictions of their transient behaviors as obtained using a dynamic sheath model for non-uniform plasmas with ion drift. It is found that the thickness of the expanding sheath edge is reduced considerably by the ion drift velocity when compared with the thickness without ion drift. Moreover, because the ion drift prevents the propagation of the rarefaction wave significantly, the phase velocity of the wave is observed to be much less than the Bohm speed. The propagating characteristics of the rarefaction wave confirm that the ion drift velocity plays an important role in the dynamic sheath expansion in non-uniform plasmas with ion drift. The results are expected to be useful in analyzing the dose and energy of implanted ions as well as understanding the sheath dynamics in a real plasma source ion implantation system in which the plasma sheath commonly evolves in a non-uniform plasma with ion drift.

Summary form only given. The authors have partially characterized an ECR (electron cyclotron resonance) etching device by using a microwave interferometer in the source region and Langmuir probes and an FTIR (Fourier transform infrared) in the downstream target region. The microwave interferometer gives a measure of the line averaged density in the source and is present for all the measurements. For a given set of conditions, the Langmuir probes are used to measure the plasma density and electron temperature radial profiles in the downstream region. The FTIR monitors the neutral gas concentration of the etching feed gas, CF/sub 4/. The interferometer is used to cross check that the conditions are similar. With the feed gas in the source region, the line averaged plasma at 10/sup 12/ cm/sup -3/, and neutral pressure below 2 mTorr, a significant fraction (80%) of the CF/sub 4/ feed gas is `burned out' by the plasma. This fraction varies with the neutral pressure, gas flow rate, and plasma density. This is consistent with a 2-D plasma-`chemistry' code developed to model this ECR system. The code predicts that varying the feed gas location should also vary the neutral composition.