Electron Trajectory Simulation Research Articles

Simulations of electron trajectories under the influence of an array of permanent magnets in a compact ion source

Trajectories of electrons emitted from a filament in a discharge ion source are computer simulated in order to investigate their behavior under the influence of different magnetic fields. These calculations allowed a better understanding of the high efficiency obtained in a recently developed compact ion source.

Study by Monte-Carlo simulation of resolution improvement by energy filtering in Bio-TEM

An electron microscope for observing biological samples at atmospheric pressure has been proposed. The major principle of the present instrument is that a biological sample is set in a conventional transmission electron microscope (TEM) beamline but it is isolated from a high vacuum by upper- and lower bulkheads. As indicated by an electron trajectory simulation, a clear image could be obtained if the bulkhead was 100 nm thick beryllium foil. In the present study, the feasibility of the image resolution improvement is examined by energy filtering of the transmitted electrons.

Femtosecond Synchroscan Streak Tube

A femtosecond synchroscan streak tube has been developed. In order to achieve femtosecond time resolution, the following conditions were adopted in the tube design: high electric field application near the photocathode, high positive voltage application to the focusing electrode to reduce photoelectron transit time spread in the focusing region, and setting the anode electric potential as low as possible to obtain a very high deflection sensitivity for sweep due to low photoelectron drift velocity between the anode and the sweeping screen. The electrode structure and the applied voltages were designed by computer simulation of electron trajectory, and the theoretical time resolution obtained is ∼520 fs. Based on the above design parameters, a prototype tube was fabricated and evaluated. A time resolution of ∼660 fs was experimentally obtained at the synchroscan frequency of ∼80 MHz.

Quantifying the cathodoluminescence generated in ZnS-based phosphor powders

Field emission displays (FEDs) are a possible alternative to other display technologies. They work similar to cathode ray tubes, but instead of an electron gun they have an array of tiny metallic tips acting as electron emitters. This array is situated in close proximity to the phosphor screen and produces light by a process of cathodoluminescence (CL). During prolonged electron beam irradiation a non-luminescent ZnO layer forms on the irradiated surface of the ZnS phosphor powder. This is due to electron-beam-stimulated surface chemical reactions between the ZnS phosphor and water vapour present in the ultrahigh vacuum environment. The low-energy electrons used in FEDs have a shallow penetration depth and, because the CL is dependent upon the energy loss in the ZnS bulk itself, growth of the ZnO layer significantly degrades the CL intensity. Energy loss profiles in the ZnS for different ZnO thicknesses were determined using Monte-Carlo electron trajectory simulations. These profiles, along with photon absorption profiles, were used to determine a curve relating the normalized CL intensity to the ZnO thickness. It was compared with experimentally measured values obtained during CL degradation experiments on standard ZnS : Cu,Al,Au phosphor powders and ZnO thickness measurements with sputter depth profiling. Assuming the presence of a 20 nm thick diffusion interface between the ZnO layer and the ZnS bulk, the theoretical and experimental results agree well. Initially the interface represents the initial stages of oxidation, but after the oxide layer starts to grow as a uniform film it represents a physical interface formed due to diffusion processes. Copyright © 2002 John Wiley & Sons, Ltd.

The influence of the Coulomb interaction effect in the electron beam on the developed resist structure for the projection lithography

A series of simulations are introduced to express the processes of the electron projection lithography quantitatively. It consists of the following three major steps: (1) electron trajectory simulation in an electron beam projection lithography optical system considering the Coulomb interaction effect among electrons and the lens aberrations, (2) electron energy deposition simulation in a resist on Si substrate, and (3) time evolution three-dimensional resist-development simulation. They are used to predict the error propagation characteristics in the sequence of lithography processes up to the three-dimensional resist structure after development. The influence of a variation in the exposure intensity on the final resist structure is demonstrated as an example.

Simulation of planar deflection systems for field emission devices

Planar electron optical systems, with emphasis on planar deflectors, were investigated by computer simulation in three dimensions (3D). The simulations were carried out using a computer program based on the surface charge density method, with a coordinate system to assist meshing of boundary elements. Simulation of electron trajectories in 3D offers the possibility to see the final spot on an anode plane, helping to optimize the electron optical design. Three types of planar deflectors, split ring, microstrip, and horizontal bars were investigated to demonstrate their capability of deflecting electrons from a field emitter or a field emitter array. Some applications for these planar deflection systems have been suggested. Simple to fabricate and easy to be integrated with a field emitter are the biggest advantages for these planar electron optical systems. With the assistance of computer simulation, a better performance from these planar systems can also be achieved.

Porous field emission devices based on polyimide membranes using diode and triode configurations

Residual gas inside field emission displays (FED) is the most important issue related to the device lifetime. Increasing the display area while maintaining the display thickness unchanged results in lifetime decrease, since the pressure gradient is fostered. Therefore, improvement of vacuum properties is a mandatory step towards large area displays. In a prior publication we have demonstrated that porous diamond membranes show good vacuum performance, while requiring low emitter switching voltage. In this work, we continue the porous membrane development by using polyimide as the base material for the membrane. The use of polyimide instead of diamond allows for easier production of large area porous FEDs. In addition, we present results of preliminary field emission experiments showing a direct correlation between the emitted current and the number of pores. This result strongly suggests that the emission sites are located at the pore edges in the polyimide membranes, similar to our observations for diamond membranes. From the theoretical point of view, we propose a new geometry, still based on the use of pores, but including a grid for triode mode operation. Finally, we present electron trajectory simulations that address some of the focusing issues in the proposed device.

Computer simulation of electron trajectories with the space charge in cascade static lens gauge

The cascade static lens gauge (CSLG) is an ionization vacuum gauge which has a very high sensitivity by oscillating electrons in the electrostatic lens system. The high sensitivity was not necessarily obtained experimentally in the condition of electric potential of electrodes in which electrons oscillate many times by the computer simulation study of electron trajectories without space charge. In this time electron trajectories have been simulated under the consideration of space charge. The space charge distribution is estimated to be proportional to the accumulated resident time of electrons in each mesh. The simulation method is proposed and typical results are shown.

Beam Characteristics of Mechanically Tunable Magnetron Injection Guns

Gyrotron has received extensive attention owing to its high-power capability, particularly when the wavelength shrinks below the millimeter-wave range. The electron beam of a gyrotron is typically generated by a magnetron injection gun (MIG), and a mechanically tunable MIG is employed to enhance the beam quality. However, a slight adjustment in the position of the center electrode of the mechanically tunable MIG can cause significant modification of the electric-field profile. Consequently, the cathode current of the mechanically tunable MIG is not only dependent on the cathode temperature, but also the relative position of center electrode. To accurately determine the beam characteristics of the mechanically tunable MIG, an improved computer program is employed to simulate electron trajectories. Simulation and measurement results indicate that although the magnetic field and compression ratio do not influence the beam current of the mechanically tunable MIG, but the beam current of the MIG depends on the cathode temperature and relative position of the center electrode. Finally, an improved mechanically tunable MIG design is developed to enhanced an beam quality while minimizing beam current variation.

Velocity ratio measurements on a 20-60 kV quasi-optical gyrotron electron beam

This paper describes experimental measurements of the velocity ratio (/spl alpha/=/spl nu//sub /spl perp////spl nu//sub /spl par//) of an annular electron beam for use in a quasi-optical gyrotron (QOG). The electron gun was originally designed for 80-kV 35-A operation, and identical guns have been used in gyrotron experiments by several research groups. A pair of capacitive probes in the drift region are used to characterize the beam over a wide range of voltage (20-60 kV) and current (0-6 A) while varying the magnetic compression and the intermediate-anode voltage. At low-beam voltage, the peak measured values of /spl alpha/ are much lower than those calculated from electron trajectory simulations. The capacitive probe results indicate that electrons reflecting from the peak of the magnetic field are important, increasing the operating voltage of the gun to 50 kV greatly improves the quality of the electron beam.

Influence of electron acceleration voltage in the cell-projection lithography system

The electron trajectory simulation, considering the Coulomb interaction effect among electrons in the beam in a given electron optical system is performed. The method to obtain the resist pattern width is introduced from the results of the simulation. The variations of the line width with the defocus, or with the exposure dose obtained by the simulation are compared with the experimental results. Finally, the exposure pattern contrast at the wafer surface is obtained as a combination of accelerating voltage and the current density of the electron beam by the simulation.

Study of the effect of layer thickness, beam energy, and metal density on the resistless silicide direct-write electron-beam lithography process for the fabrication of nanostructures

To overcome the limitation of resists in electron beam lithography, a resistless electron beam lithography technique was recently developed. In the silicide direct-write electron-beam lithography process (SiDWEL), a thin metallic film is deposited on a silicon surface. A low-energy (<3 keV) electron beam is then used to enable the intermixing of the metal and the silicon layers through thermal effects. A chemical etch is then used to remove the unexposed metal regions. Thermal calculations are performed using a Monte Carlo simulation of electron trajectories and are correlated with experiments using Ni as the thin metallic film. A comparison of the doses required for the formation of several metals is also done. Results show that the SiDWEL process is possible when the electrons lose all their energy in a layer thickness comparable to the phonon mean free path. Finally, experiments are performed using multilayer samples to form silicide structures.

電子ビーム照射に伴う絶縁物帯電のシミュレーション

A simulation model is developed to express the time dependent charging-up process of insulators under electron beam irradiation. The electron deposition and the energy deposition distributions produced by the incident electron beam are calculated by a Monte Carlo simulation of electron trajectories. The trajectories of slow secondary electrons and the Auger electrons are calculated considering the electric field. The electron beam induced conduction is calculated with taking into account the surface boundary of the specimen. The time-dependent spatial distributions of the charge accumulated, the electric potential built, and the backscattered secondary yield by the Gaussian-shaped electron beam irradiation are obtained. The positive and negative charging processes of a bulk PMMA by 1 keV and 5 keV electron beam irradiation, respectively are expressed.

Monte Carlo simulation on the electron beam incident angle with spherical particles applied to the energy loss in ZnS phosphor powders

During electron beam irradiation of ZnS phosphor powders, a non-luminescent ZnO layer is formed on the powder due to electron-beam-stimulated surface reactions. As the thickness of the oxide layer increases, the energy loss in the ZnS bulk decreases with a subsequent degradation in cathodoluminescence (CL). Owing to the morphology of the phosphor powder, growth of the oxide layer leads to varying effective ZnO thicknesses. In this paper the effect of the interaction between the electron beam and the ZnO/ZnS powder particles on the energy loss in the ZnS is studied using Monte Carlo simulation techniques. In the simulation, an electron beam with a certain profile is spread according to a Gaussian distribution function over the phosphor particles. These particles consists of both flat and spherical-shaped grains. The incident angle between the electron beam and the particle's surface is determined by using simple vector analysis. If the beam contains a sufficient number of electrons, a distribution of incident angles can be obtained. It was found that this distribution is independent upon the profile of the electron beam as long as the full width at half-maximum (FWHM) beam diameter is larger than the size of the irradiated particles. Using these results a comparison was made between the energy loss in the ZnS particles as a function of the ZnO thickness with and without the angular distribution. A publicly available Monte Carlo code for electron trajectory simulations in solids, called CASINO, was used for this purpose. When the distribution is considered, there is a decreased energy loss in the ZnS bulk. Because the light generation during electron beam irradiation is proportional to this energy loss, the morphology of the phosphor powder also contributes to the CL degradation. Copyright © 2000 John Wiley & Sons, Ltd.

Characteristic Variation of Exposure Pattern in Cell-Projection Electron-Beam Lithography

The characteristic variation of an exposure pattern with various optical parameters is calculated in a typical cell-projection lithography system by using an electron trajectory simulation. The Coulomb interaction effect among electrons in the beam is calculated which determines the pattern blur at the wafer surface. The acceleration voltage is 50 kV, the exposure pattern is a series of line and space, and the field size is 5 µm×5 µm. As the beam current density varies from 2 to 13 A/cm2 at the specimen surface, the pattern blur is evaluated in parameters of (1) the probability of electrons to be in the designed pattern, (2) the full width at half maximum and (3) the contrast of the electron density distribution at the wafer surface, as the line width varies from 0.13 to 0.2 µm.

Dynamic Simulation of Electron-Beam-Induced Chargingup of Insulators

The time-dependent charging process of an insulating specimen under electron beam irradiation is calculated by taking into account the charge continuity equation, the Poisson equation, Ohm's law and the electron-beam-induced conductivity. The energy and the charge deposition distributions are calculated by the Monte Carlo simulation of electron trajectories taking into account the electric field distribution in the specimen. The time-dependent charge and potential distributions are obtained in the present calculation. As the electron beam energy is 10 keV and the specimen is a 1-mm-thick poly-methyl-methacrylate (PMMA) wafer, the surface potential increases to a positive value at first, and then becomes a negative one. This charging process agrees with experimental findings.

Simulation of Energy Deposition in E-Beam Irradiated Polymers

IntroductionRadiation curing of polymers, by UV light or E-beam, is of growing interest in the state-of-the-art industry. Radiation curing is solvent free and leads to important energy and money savings. Improved materials characteristics, better control of doses and more accurate application allow radiation curing to take a big part of the surface coating industry and, for the last ten years, the polymeric and microelectronics (lithography) industry. But behaviors of polymers under radiant source is often empirically characterized. In this abstract, it is proposed a way to obtain accurate energy distribution from Monte Carlo simulations. For bulk or thin polymeric films, it is therefore essential to simulate energy distribution, in order to foresee what happens in the sample.ResultsCASINO is a single scattering MonteCarlo Simulation of electroNtrajectory in sOlid -particularly designed for low energy beam interaction in a bulk and thin foil.

Optics and design of the fringe field monochromator for a Schottky field emission gun

Abstract For the improvement of high-resolution electron energy loss spectroscopy a new electron source monochromator, based on the Wien filter principle, is presented. In the fringe field monochromator the electric and magnetic filter fields are tightly enclosed by field clamps to satisfy the Wien condition, E=vB. The whole monochromator including the 150 nm energy selection slits (Nanoslits) is positioned in the gun area. Its total length is only 42 mm. Using electron trajectory simulation through the filter fields the dispersion and aberrations are determined. The parasitic astigmatism of the gun lens needs to be corrected using an electrostatic quadrupole field incorporated in the filter. Estimations of the influence of filter electrode misalignment show that at least six filter electrodes must be used to loosen the alignment demands sufficiently. Using theoretical estimations of the Coulomb interaction the final energy resolution, beam brightness and current are predicted. For a Schottky field emission electron gun with typical brightness of 108 A/sr m2 V the monochromator is expected to produce a 50 meV 1 nA beam with a brightness of 107.

部分一括露光方式電子ビームリソグラフィ用シミュレータの開発

We developed a Monte Carlo simulation of electron trajectories in a cell-projection electron beam lithography optical system. In the simulation the emission angle, the velocity and the emission timing of electrons from the electron gun are taking into account. The Coulomb interaction among electrons in the beam is calculated. There are five lenses in the system. Assuming that the stencil mask pattern is a series of line-and-space, the exposure pattern is obtained when the current densities are 5, 10, and 13 A/cm2 at the specimen surface. The probability of electrons to enter the designed pattern as a function of the defocusing distance are calculated. The best refocus length is obtained to give the maximum probability in the designed pattern at the specimen surface.

Analysis of the Image Formation Mechanism on High Energy Scanning Electron Microscopy

The image formation mechanism of the high-energy scanning electron microscopy is analyzed by a Monte Carlo simulation of electron trajectories in a specimen. It is shown that only secondary electrons which are produced directly by incident primary electrons carry information of the three-dimensional hole structure at the specimen surface. The major contribution to the signal is due to the secondary electrons emitted from the side-wall of the hole. Even if the hole is covered by a 30-nm-thick carbon film, the edge contrast of the hole is appreciable as the primary electron is incident normal to the surface. If there is an aperture in the film covering the hole, secondary electrons from the side-wall of the hole contribute to the signal.