Diagnostics of Low-Pressure Oxygen RF Plasmas and the Mechanism for Polymer Etching: A Comparison of Reactive Sputter Etching and Magnetron Sputter Etching

We have compared low-pressure oxygen RF plasmas and the etching of photoresist in a reactive sputter etch reactor and in a magnetron etch reactor using Langmuir probe, optical emission actinometry, and mass spectrometry measurements. The Langmuir probe data allow the determination of the plasma ion density and electron temperature, and thus the ion flux onto the substrate. The optical data yield information on the presence of O atoms and O2+ ions. Stable reactant and product species are monitored with a mass spectrometer. The main difference between the two reactors is that in magnetron sputter etching (MSE), the ion flux to the substrate is about an order of magnitude higher, under comparable plasma conditions, than in reactive sputter etching (RSE). This accounts for the higher etch rate in MSE. However, the etch yield per ion is higher in RSE because of the higher ion energy. Etch rates correlate neither with the ion flux to the substrate nor with the density of O atoms in the plasma, but change in parallel with the consumption of reactant gas. We conclude that in etching a polymer in a low-pressure oxygen plasma, the main neutral reactant species are O2 molecules, and an important role of the ions is to remove reaction products from the substrate surface.     

Two-dimensional direct simulation Monte Carlo (DSMC) of reactiveneutral and ion flow in a high density plasma reactor

We present a two dimensional direct simulation Monte Carlo (DSMC) study of the rarefied reactive flow of neutrals and ions in a low pressure inductively coupled plasma reactor. The spatially-dependent rate coefficients of electron impact reactions and the electrostatic field were obtained from a fluid plasma simulation. Neutral and ion etching of polysilicon with chlorine gas was studied with emphasis on the reaction uniformity along the wafer. Substantial gradients in total gas density were observed across the reactor invalidating the commonly made assumption of constant gas density. The flow was nonequilibrium with differences in the species translational temperatures, and 100 K temperature jumps near the walls. When etching was limited by ions the etch rate was highest at the wafer center. When etching was limited by neutrals, the etch rate was highest at the wafer edge. In such case, the etch uniformity changed significantly depending on the reactivity of the ring surrounding the wafer. The ion angular distribution was several degrees off normal and it was different at the wafer edge compared to the rest of the wafer     

Direct simulation Monte Carlo analysis of flows and etch rate in aninductively coupled plasma reactor

The direct simulation Monte Carlo (DSMC) method was employed to predict the etch rate distribution on Si wafer. The etchant is assumed to be Cl. The production rate of Cl due to electron impact was obtained separately by preprocessing an inductively coupled chlorine plasma by use of the particle-in-cell/Monte Carlo method. Under the condition of constant total pressure, the etch rate increases with the mass flow rate of source gas Cl₂. The density of the etch product SiCl₂ rapidly decreases with increasing the flow rate. The density of the etchant hardly depends on the flow rate. The recombination 2Cl→Cl₂ on the inner walls of etching apparatus has a large effect on the etch rate; recombination probability of 0.1 results in 50% reduction of the etch rate. The etch rate distribution becomes more uniform when the reaction probability at the wafer surface is reduced     

Energy transport in plasma etching of nanoporous dielectric materials

A Monte Carlo routine was developed to simulate the motion and energetics of ions in the pores of a xerogel material under plasma etching conditions. The simulation included the effects of an applied electric field and input conditions for the pore as a function of pressure and applied voltage in the plasma reactor. We were interested in the ion energy in a pore, the ion penetration depth and the effect of ion energy on etching.At low pressures the nanoporous material etches faster than dense silicon dioxide. This is to be expected given the decrease in density and increase in surface area that arises due to the porosity. However, as the pressure is increased, the etch rate decreases dramatically and, eventually, the dense oxide may etch faster than the porous material. CHF₃ was used as the etchant gas and, for this gas, we believe this behavior to be controlled by the ion energy and energy transport in the pores of the xerogel material. As the pressure in the plasma reactor is increased, the incoming ions switch over from etching activation to polymerisation activation. This agrees with the observed crossover in etch rate seen experimentally and with the cessation in etching as pressure is increased. The switch is affected by pore roughness and correlates with the average ion energy in the pore.     

Two-dimensional simulation of polysilicon etching with chlorine ina high density plasma reactor

A two-dimensional fluid simulation of polysilicon etching with chlorine in an inductively-coupled high density plasma source is presented. A modular approach was used to couple in a self-consistent manner the disparate time scales of plasma and neutral species transport. This way, complex plasma chemical reactions (involving electrons, ions and neutrals) as well as surface chemistry can be included in the simulation, The power deposited into the plasma was calculated by an electromagnetics module which solves Maxwell's equations. The power deposition was used in the electron energy module to find the electron temperature and the rate coefficients of electron-impact reactions. These were in turn used as source terms in separate neutral and charged species transport modules. By iterating among the modules, a self-consistent solution was obtained. Quantities of interest, such as power deposition, species density and flux, and etch rate and uniformity were thus calculated, As power deposition was increased, the electron density increased linearly, the plasma became less electronegative, the degree of gas dissociation increased, and the plasma potential remained constant. The radial uniformity of the Cl atom flux was better than that of the ion flux. The reactivity of the wafer as compared to that of the surrounding electrode surface significantly affected the etch uniformity, despite the low pressure of 10 mtorr     

Investigations on the effect of process parameters on the growth of vertically oriented graphene sheet in plasma-enhanced chemical vapour deposition system

A multistage numerical model comprising the plasma kinetics and surface deposition sub-models is developed to study the influence of process parameters, namely, total gas pressure and input plasma power on the plasma chemistry and growth characteristics of vertically oriented graphene sheets (VOGS) grown in the plasma-enhanced chemical vapour deposition system containing the Ar + H₂ + C₂H₂ reactive gas mixture. The spectral and spatial distributions of temperature and number densities, respectively, of plasma species, that is, charged and neutral species in the plasma reactor, are examined using inductively coupled plasma module of COMSOL Multiphysics 5.2 modelling suite. The numerical data from the computational plasma model are fed as the input parameters for the surface deposition model, and from the simulation results, it is found that there is a significant drop in the densities of various plasma species as one goes from the bulk plasma region to the substrate surface. The significant loss of the energetic electrons is observed in the plasma region at high pressure (for constant input power) and low input power (for constant gas pressure). At low pressure, the carbon species generate at higher rates on the catalyst nanoislands surface, thus enhancing the growth and surface density of VOGS. However, it is found that VOGS growth rate increases when input plasma power is raised from 100 to 300 W and decreases with further increase in the plasma power. A good comparison of the model outcomes with the available experimental results confirms the adequacy of the present model.     

Model for ion-initiated trench etching

We model the plasma etching of trenches by Langmuir kinetics for neutral molecules and bombarding ions. The parallel combination of an isotropic etch rate for the neutrals and an anisotropic etch rate for the ions gives an effective etch rate. The ion etch rate is proportional to the normal surface component of the ion energy flux. An approximate analytical expression for the composite etch rate offers a new approach to the computation of etch profiles for these mixed systems. Etch profiles are displayed for three cases: the nearly ion flux-limited regime, an intermediate case, and the nearly neutral-flux limited regime for the trench bottom. The numerical calculation of the etch profiles follows from the integration of three characteristic strip equations which are nonlinear first-order ordinary differential equations (ODE's)     

Dry etching of ITO by magnetic pole enhanced inductively coupled plasma for display and biosensing devices

The dry etching of indium tin oxide (ITO) layers deposited on glass substrates was investigated in a high density inductively coupled plasma (ICP) source. This innovative low pressure plasma source uses a magnetic core in order to concentrate the electromagnetic energy on the plasma and thus provides for higher plasma density and better uniformity. Different gas mixtures were tested containing mainly hydrogen, argon and methane. In Ar/H₂ mixtures and at constant bias voltage (−100 V), the etch rate shows a linear dependence with input power varying the same way as the ion density, which confirms the hypothesis that the etching process is mainly physical. In CH₄/H₂ mixtures, the etch rate goes through a maximum for 10% CH₄ indicating a participation of the radicals to the etching process. However, the etch rate remains quite low with this type of gas mixture (around 10 nm/min) because the etching mechanism appears to be competing with a deposition process. With CH₄/Ar mixtures, a similar feature appeared but the etch rate was much higher, reaching 130 nm/min at 10% of CH₄ in Ar. The increase in etch rate with the addition of a small quantity of methane indicates that the physical etching process is enhanced by a chemical mechanism. The etching process was monitored by optical emission spectroscopy that appeared to be a valuable tool for endpoint detection.     

Effects of nitrogen addition on methane-based ECR plasma etching of gallium nitride

The effects of nitrogen addition on methane-based ECR plasma etching of GaN were studied. The etch rate 30 nm/min and r.m.s. roughness 2.6 nm were obtained when the GaN sample was etched by a methane-based gas mixture without N₂. The addition of N₂ gas resulted in a decrease of etch rate and a smoother etched surface. The r.m.s. roughness became less than 0.4 nm even only 1.5 sccm N₂ gas was added to the mixture. In situ XPS measurements showed that, without N₂, heavy N-depletion took place on the etched surface, resulting in appearance of Ga metal on the surface. In contrast, the loss of N was compensated when the N₂ gas was added, and the etched surface approached the stoichiometric one with the increase of N₂ gas flow. This suppression of preferential loss of N was considered to be the main reason that improved the etched surface morphology.     

充中性气体相对论返波振荡器的粒子模拟研究

 用PIC粒子模拟方法研究了充中性气体相对论返波管的物理机制，成功模拟了电子束碰撞充入返波管中的中性气体电离产生等离子体的过程，在电子束传输的路径上形成离子通道，有效中和电子束径向空间电荷力，有利于电子束的传输及束波相互作用产生微波。增加中性气体密度，返波管的输出频率明显上移，其辐射的功率和效率比相同的真空器件也有明显的提高。     

Etching and oxidation of InAs in planar inductively coupled plasma

The surface of InAs (1 1 1)A was investigated under plasmachemical etching in the gas mixture CH₄/H₂/Ar. Etching was performed using the RF (13.56 MHz) and ICP plasma with the power 30–150 and 50–300 W, respectively; gas pressure in the reactor was 3–10 mTorr. It was demonstrated that the composition of the subsurface layer less than 5 nm thick changes during plasmachemical etching.A method of deep etching of InAs involving ICP plasma and hydrocarbon based chemistry providing the conservation of the surface relief is proposed. Optimal conditions and the composition of the gas phase for plasmachemical etching ensuring acceptable etch rates were selected.     

Surface-science aspects of plasma-assisted etching

Plasma-assisted etching methods have been used in the manufacture of integrated circuits for more than 10 years and yet the surface-science aspects of this technology are poorly understood. The chemistry must be such that the reactive species generated in the plasma react with the surface being etched to form a volatile product. The chemistry is usually dominated by atoms, molecular radicals and low-energy (20–500 eV) positive ions. In microstructure fabrication, the positive ions are accelerated from the plasma towards the etched surface arriving essentially at normal incidence. Thus, the bottom surface of a very small feature being etched is subjected to both energetic ions and reactive neutral species, whereas the sidewalls of the feature are exposed to reactive neutral species only. The role of the energetic ions is primarily to accelerate the reaction between the neutral species and the etched surface (i.e., accelerate the etch rate), thereby reducing the steady-state top-monolayer coverage of the etching species on the etched surface. On the sidewalls, however, the reacting-species coverage is a saturation coverage. The present understanding of some of the surface-science aspects of this complex environment will be summarized, often using the Si-F system as an example, and some phenomena which are not well understood will be described.     

Dry etching characteristics of GaN using Cl2/BCl3 inductively coupled plasmas

ICP power/RF power, operating pressure, and Cl₂/BCl₃ gas mixing ratio are altered to investigate the effect of input process parameters on the etch characteristics of GaN films. The etch selectivity of GaN over SiO₂ and photoresist is studied. Although higher ICP/RF power can obtain higher GaN/photoresist etch selectivity, it can result in faceting of sidewall and weird sidewall profile due to photoresist mask erosion. Etch rates of GaN and SiO₂ decrease with the increase of operating pressure, and etch selectivity of GaN over SiO₂ increases with the increasing operating pressure at fixed ICP/RF power and mixture component. The highest etch selectivity of GaN over SiO₂ is 7.92, and an almost vertical etch profile having an etch rate of GaN close to 845.3 nm/min can be achieved. The surface morphology and root-mean-square roughness of the etched GaN under different etching conditions are evaluated by atomic force microscopy. The plasma-induced damage of GaN is analyzed using photoluminescence (PL) measurements. The optimized etching process, used for mesa formation during the LED fabrication, is presented. The periodic pattern can be transferred into GaN using a combination of Cl₂/BCl₃ plasma chemistry and hard mask SiO₂. Patterning of the sapphire substrate for fabricating LED with improved extraction efficiency is also possible using the same plasma chemistry.     

Particle‐in‐cell/Monte Carlo simulation of electron and ion currents to cylindrical Langmuir probe

Electron and ion currents to a cylindrical Langmuir (electrostatic) probe were calculated using the particle‐in‐cell/Monte Carlo (PIC/MC) self‐consistent simulation for a neutral gas in the pressure range 2–3,000 Pa. The simulation enables us to calculate the probe currents even at high neutral gas pressures when the collisions of collected charged particles with neutral gas particles near the probe are important. The main aim of this paper is the calculation of probe currents at such high gas pressures and the comparison of the results with experimentally measured probe currents. Simulations were performed for two cases: (a) probes with varying radii in a non‐thermal plasma with high electron temperature at low neutral gas pressure of 2 Pa (in order to verify the correctness of our simulations), and (b) probe with the radius of 10 μm in the afterglow plasma with low electron temperature and a higher neutral gas pressure (up to 3,000 Pa). The electron probe currents obtained in case (a) show good agreement with those predicted by the orbital motion limited current (OMLC) theory for probes with radii up to 100 μm for the given plasma conditions. At larger probe radii and/or at higher probe voltages, the OMLC theory incorrectly predicts too high an electron probe current for the plasma parameters studied. Additionally, a formula describing the spatial dependence of the electron density in the presheath in the collisionless case is derived. The simulation at higher neutral gas pressures, i.e. case (b), shows a decrease of the electron probe current with increasing gas pressure and the creation of a large presheath around the probe. The simulated electron probe currents are compared with those of measurements by other authors, and the differences are discussed.     

Fluorine etching on the Si(111)-7×7 surfaces using fluorinated fullerene

Fluorine etching on the Si(1 1 1)-7×7 surfaces using fluorinated fullerene molecules as a fluorine source has been investigated. At room temperature, adsorbed fluorinated fullerene molecules reacted with the Si(1 1 1)-7×7 surface to create a localized distribution of fluorine on the surface. Nanoscale etch pits were created by annealing at 300 °C, due to the adsorption of the fluorine localized around the C₆₀F_x molecules. Annealing at 400 °C resulted in the delocalized fluorine distribution on the surface and healing of the etch pits, due to the enhancement of the diffusion of both the fluorine and silicon atoms. Subsequent annealing at 500 °C led to desorption of SiF₂ reactants formed on the surface. The fluorine diffusion process was found to be an elemental process in the etching because the diffusion of adsorbed fluorines is a key for the formation of the SiF₂ species and their subsequent desorption.     

Numerical simulation on the influence of water spray in thermal plasma treatment of CF4 gas

Nitrogen thermal plasma generated by a non-transferred DC arc plasma torch was used to decompose tetrafluoromethane (CF₄). In the thermal decomposition process, water was used as a chemical reactant source. Two kinds of water spray methods were compared: water spray directly to the arc plasma flame and indirectly to the reactor tube wall. Although the same operating conditions of input power, waste gas, and sprayed water flow rate were employed for each water spray methods, a relatively higher decomposition rate was achieved in the case of water spray to the reactor wall. In order to investigate the effects of water spraying direction on the thermal decomposition process, a numerical simulation on the thermal plasma flow characteristics was carried out considering water injection in the reactor. The simulation was performed using commercial fluid dynamics software of the FLUENT, which is suitable for calculating a complex flow. From the results, it was revealed that water spray to the reactor wall and use of a relatively small quantity of water are more effective methods for decomposition of CF₄, because a sufficiently high temperature area and long reaction time can be maintained over large area.     

Influence of a centered dielectric tube on inductively coupled plasma source: Chamber structures and plasma characteristics

A high-density RF ion source is an essential part of a neutral beam injector. In this study, the authors attempt to retrofit an original regular RF ion source reactor by inserting a thin dielectric tube through the symmetric axis of the discharge chamber. With the aid of this inner tube, the reactor is capable of generating a radial magnetic field instead of the original transverse magnetic field, which solves the E × B drift problem in the current RF ion source structure. To study the disturbance of the dielectric tube, a fluid model is introduced to study the plasma parameters with or without the internal dielectric tube, based on the inductively coupled plasma(ICP) reactor. The simulation results show that while introducing the internal dielectric tube into the ICP reactor, both the plasma density and plasma potential have minor influence during the discharge process, and there is good uniformity at the extraction region. The influence of the control parameters reveals that the plasma densities at the extraction region decrease first and subsequently slow down while enhancing the diffusion region.     

Dry etching characteristics of GaN for blue/green light-emitting diode fabrication

The etch rates, surface morphology and sidewall profiles of features formed in GaN/InGaN/AlGaN multiple quantum well light-emitting diodes by Cl₂-based dry etching are reported. The chlorine provides an enhancement in etch rate of over a factor of 40 relative to the physical etching provided by Ar and the etching is reactant-limited until chlorine gas flow rates of at least 50 standard cubic centimeters per minute. Mesa sidewall profile angle control is possible using a combination of Cl₂/Ar plasma chemistry and SiO₂ mask. N-face GaN is found to etch faster than Ga-face surfaces under the same conditions. Patterning of the sapphire substrate for improved light extraction is also possible using the same plasma chemistry.     

Deep SiC etching with RIE

SiC is currently an important topic in power devices. This new technology leads to lower power losses, faster switching, and higher working temperature. The design of SiC power devices requires the integration of edge termination techniques to obtain a high blocking voltage. The mesa structure approach is one well-established method. It could be used alone or in combination with a Junction Termination Extension (JTE). The mesa consists of a structure that removes material around the pn-junction. Due to the strong Si–C bonds, conventional chemical–wet etching solutions are inefficient on SiC, so plasma methods are required to etch SiC.The presented work is based on the use of an RIE reactor with an SF₆/ O₂ plasma. Its geometry structure and parameters were optimized. An etch rate of 0.35 μm/min was obtained without any trenching phenomenon. Trenches deeper than 10 μm deep were realized with a nickel etching mask that shows a high selectivity. AFM analysis revealed an etched surface as smooth as the initial one.     

Photoemission study of fluorination atmospheric pressure plasma processes on EPDM: Influence of the carrier and fluorinating gas

Fluorination plasma treatments at atmospheric pressure were used to modify the surface composition of EPDM elastomer. In this study, two different precursors (CF₄ and SF₆) and two carrier gases (He and Ar) were used for the surface modification of EPDM elastomer. The surface modifications were studied by means of X-ray photoelectron spectroscopy. We have observed a strong influence of the gas selection on the extent of the surface modification induced with these treatments. In general terms, the use of CF₄ generates a higher concentration of fluorine in the elastomer surface. On the other hand, the use of He as carrier gas also increases the effectiveness of the modification process. The fluorine uptake varies between 2 and 13%, although the formation of fluorine-containing functional groups was detected when the amount of fluorine on the surface exceeded 7%. After all treatments, an important oxygen uptake was observed, with amounts three or four times higher than the untreated elastomer.