Electron Trapping Effect Research Articles

On the possibility of current drive in tokamaks by Bernstein modes

Current drive in a tokamak by electrostatic electron Bernstein waves is investigated. It is shown that a sufficiently large refractive index of these waves may result in regimes of rf power dissipation in which the effect of electron trapping into “banana” orbits will be essentially diminished. Besides, a stronger (as compared with O- and X-modes) absorption of electrostatic electron Bernstein modes results in an rf power input into more energetic electrons which is also favourable for the increase of the current drive efficiency.

Current drive with fast waves, electron cyclotron waves, and neutral injection in the DIII-D tokamak

Current drive experiments have been performed on the DIII-D tokamak using fast waves, electron cyclotron waves, and neutral injection. Fast wave experiments were performed using a 4-strap antenna with 1 MW of power at 60 MHz. These experiments showed effective heating of electrons, with a global heating efficiency equivalent to that of neutral injection even when the single pass damping was calculated to be as small as 5%. The damping was probably due to the effect of multiple passes of the wave through the plasma. Fast wave current drive experiments were performed with a toroidally directional phasing of the antenna straps. Currents driven by fast wave current drive (FWCD) in the direction of the main plasma current of up to 100 kA were found, not including a calculated 40 kA of bootstrap current. Experiments with FWCD in the counter current direction showed little current drive. In both cases, changes in the sawtooth behavior and the internal inductance qualitatively support the measurement of FWCD. Experiments on electron cyclotron current drive have shown that 100 kA of current can be driven by 1 MW of power at 60 GHz. Calculation with a Fokker-Planck code show that electron cyclotron current drive (ECCD) can be well predicted when the effects of electron trapping and of the residual electric field are included.

Evidence of electron trapping in undoped hydrogenated amorphous silicon (a-Si:H) Schottky barriers by the surface photovoltage (SPV) technique — A study of stability in a-Si:H solar cells

Abstract Surface photovoltage (SPV) measurements on a-Si:H/metal Schottky devices under forward bias are presented. The space-charge density decreases with applied forward bias while both the space-charge width and the apparent diffusion length are essentially constant. These are all attributed to electron trapping in Schottky barrier region. Furthermore, we explore electron trapping effect in a-Si:H on light-induced degradation of a-Si:H solar cells.

Enhanced electron trapping in reoxidized nitrided oxide dielectric at 77 K

The results of experiments on the electron trapping effect in ONO devices at cryogenic temperature are presented and compared with results for SiO/sub 2/ devices. Various types of ONO devices are used in the experiment in order to determine the mechanism of enhanced electron trapping phenomena in these devices. >

Constant capacitance technique to study electrical instabilities in InP MIS provided by PECVD silicon nitride

The charge trapping behaviour in silicon nitride/InP metal insulator semiconductor (MIS) capacitors is investigated by a constant capacitance technique at room temperature. Silicon nitride films were formed by a conventional plasma enhanced chemical vapor deposition (PECVD) technique where MIS structures were grown simultaneously on n- and p-type InP to observe the effect of electron and hole trapping, respectively. We have found that the carrier trapping by direct tunneling near the silicon nitride/InP interface is dominant under low insulator fields for both n- and p-type samples. The amount of injected charges was similar for both types of MIS prepared under same condition. This behaviour is attributed to the amphoteric nature of silicon dangling bonds. The injection of carriers to the bulk of the silicon nitride assisted by the electric field occurs at high insulator fields. This phenomena is apparent particularly for MIS diodes prepared with a smaller ammonia to silane partial pressure ratio in deposition. It is also confirmed that the trap density near the silicon nitride/InP interface is about one order of magnitude higher than that of the silicon nitride bulk due to the initial transient phenomena of the PECVD process.

Influence of various physics phenomena on fast wave current drive in tokamaks

The authors present an analysis of the influence of various physics phenomena (which are not necessarily independent) on current drive performance in tokamaks. Such phenomena include diffraction and other non-geometrical optics processes, k|| modification, and multiple-pass absorption as well as antenna characteristics such as recessed cavity and septa geometry, poloidal extent and poloidal location of the current straps. The two-and-one-half-dimensional (21/2-D) full-wave code PICES is used for modelling ion cyclotron resonance heating and current drive. PICES is based on poloidal mode and reduced-order expansions. By 21/2-D, we mean that 3-D wave fields are calculated in axisymmetric geometry (2-D solution domain - ρ,θ), while the correct toroidal dependence of the antenna source currents is obtained from a 2-D (ρ ,ϕ) recessed antenna code. The calculation includes the poloidal and toroidal structure of the antenna, modification of the k|| spectrum due to the poloidal magnetic field and a complete solution for E||. A semi-analytic model for current drive, including trapped electron effects, is employed. These calculations are used extensively to model fast wave current drive in DIII-D, ITER and other tokamaks

Stimulated Raman scattering: Action evolution and particle trapping via Euler–Vlasov fluid simulation

The Manley–Rowe action relations and trapped-electron orbit theory have been applied to the problem of stimulated Raman scattering (SRS) in the forward direction, as simulated by an Euler–Vlasov fluid (EVF) code in periodic geometry. After a slight initial delay, the Vlasov results agree well with three-oscillator theory (but with damping larger than the Landau linear value). The action sum of the pump and the scattered SRS wave was well conserved, while the action loss between the pump and plasma wave was shown to be well accounted for in detail by examining the electron energy for electrons above the lower trapping boundary (separatrix) in x-px phase space. However, unlike the three-oscillator model, the action sum of the pump and plasma wave was observed to actually increase for a short time because of retrieval of apparently lost action from the trapped-electron effect in regenerating the plasma wave. Thus the utility was clearly demonstrated of using action conservation to separate the effects of three-wave interaction from those due to nonlinear particle–wave interaction. The EVF code also shows clearly the fine details of layering and folding seen in x-px phase space due to electron trapping, detrapping, and retrapping. [These details would have been invisible in the usual particle-in-cell (PIC) code because of graininess at low phase space densities.]

Physical model for characterizing and simulating a FLOTOX EEPROM device

A physical model has been developed to analyze the dynamic characteristics of a FLOTOX EEPROM device. The effects of the structural parameters such as the area and thickness of the tunneling-oxide and interpoly-oxide layers are characterized by a coupling ratio to describe the discrete programming or erasing operation. The physical parameters including the electron trapping and positive-charge generation effects are used to describe the endurance and retention operations of an EEPROM device. Computer simulations based on this model have been performed to analyze the operations of an EEPROM device, including the effects of three different programming/erasing input voltage waveforms (pulse, exponential rise and triangular). A method for protecting an EEPROM device from overshooting or undershooting during programming or erasing operation is proposed. Therefore, the proposed model can be used as a computer-aided-design (CAD) tool for device design and an efficient simulation tool for describing the dynamic operation and reliability of an EEPROM device.

Effects of plasma turbulence on electron collection by a high‐voltage spherical probe in a magnetized plasma

We study the effects of electron trapping and plasma turbulence on the current‐voltage relation for a positively charged high‐voltage spherical probe in a magnetized plasma. In a force field consisting of an attractive radial electric field and a uniform magnetic field, the electric field provides a potential well in which electrons can be bound if they have insufficient energy to escape and cannot be captured by the sphere (by virtue of angular momentum conservation and/or the effects of the magnetic field). Electrons entering such a well from infinity may become trapped if they lose energy as a result of some scattering process acting within the well, such as inelastic collisions with neutral atoms or wave‐particle interactions due to plasma collective effects. When the neutral density is low enough so that ionization can be ignored, our results indicate that the sheath structure and the current‐voltage relation of a charged sphere configured as in the originally planned SPEAR I experiment (with an operating plasma contactor) would be substantially influenced by the combined action of trapped electrons and plasma instabilities. Our method is a modification of Linson's (1969) method in which potential contours are taken to be elliptical rather than cylindrical, space charge is not assumed to be constant within the sheath, and the controlling parameter is a scattering rate. We find reduced levels of current collection relative to Linson's model. Further, variation of the scattering parameter over the whole range of physically realizable values results in a range of predicted current collection values which is a small fraction of the interval between the Parker‐Murphy and Langmuir‐Blodgett limits.

Solution of three-dimensional Fokker-Planck equations for tokamak plasmas using an operator splitting technique

A two-stage operator splitting algorithm has been used to time advance the three-dimensional Fokker-Planck equation for the electron distribution function in a tokamak. In the first stage, derivatives in the two velocity-space dimensions are advanced together implicitly. Derivatives in the remaining (spatial) dimension are advanced in the second stage. The treatment is restricted to problems which are diagonal in the spatial dimension. The algorithm gives accurate solutions in modest CPU times unlike alternative schemes considered. We described the technique which includes features that accomodate the internal boundary conditions arising from trapped electron effects.

Trapped-electron effects on time-independent negative-bias states of a collisionless single-emitter plasma device: Theory and simulation

Time-average values from particle simulations of a collisionless, single-emitter plasma device modeling single-ended Q machines or thermionic converters with a negatively biased collector are presented. These results quantitatively confirm the predictions of collisionless, kinetic plane-diode theory for spatial potential profiles that decrease monotonically. However, simulations of negative-bias potential profiles with a single internal maximum differ significantly from previous theoretical predictions which assumed electron phase space to have either (i) no trapped electrons or (ii) trapped electrons isothermal with the passing electrons. A more general class of trapped-electron model distributions is introduced from which new equilibrium potential values can be recovered that closely match the simulations. These simulations clearly demonstrate the sensitive role that trapped electrons play in shaping the potential profiles of the equilibrium (or slowly evolving) states of the simulated systems. The trapped-electron distributions in these simulations are themselves shown to be controlled critically by fluctuations whose levels are varied by the choice of particle injection scheme. These effects, although found and discussed here in the context of a particular model, are believed to be important in many bounded plasma systems where electrons can be trapped in potential wells.

Dissipative trapped electron effects on toroidal ηi-mode stability

The stability properties of ion temperature gradient driven modes including the effect of dissipative trapped electrons is analyzed based on a toroidal two-fluid plasma model. It is shown that for finite trapping the ηi-mode stability thresholds are discontinously shifted compared to the case without trapping. Furthermore the thresholds are not sensitive to the magnitude of the trapping. A stability requirement on the density scale length Ln which has to be larger than about half the major radius R and an upper and a lower ηi stability threshold which grow linearly as ± 2Ln/R are derived.

High-field electron trapping and detrapping characteristics in thin SiO xN y films

The electron trapping at high-field and detrapping in thin thermally nitrided silicon oxide (SiOxNy) films were studied in the present work. For lightly nitrided oxide, the effective electron trap concentration was found to increase rapidly with the nitridation duration. In addition, the detrapping ratio of trapped electrons is almost zero and independent of the detrapping time and the applied field strength up to 8 MV/cm. This observation indicates that the increased traps have deeper energy levels. For heavily-nitrided sample, the surface trap concentration decreases and the energy levels of these traps become shallower. The detrapping ratio therefore increases and is an exponential function of the field strength and detrapping time. The detrapping ratio remains unchanged if no external field is applied.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Time-dependent hole and electron trapping effects in SIMOX buried oxides

The back-channel threshold shift associated with the buried oxide layers of separation by implanted oxygen (SIMOX) and zone-melted recrystallization (ZMR) field-effect transistors (FETs) was measured following pulsed irradiation as a function of temperature and back-gate bias using a fast time-resolved I-V measurement technique. The SIMOX FETs showed large initial negative voltage shifts at 0.2 ms after irradiation followed by temperature- and bias-dependent additional negative shifts to 800 s. Analysis and modeling of the results indicated: (1) efficient deep trapping of radiation-generated holes in the bulk of the oxide, (2) substantial initial trapping of radiation-generated electrons in the oxide, and (3) rapid removal of the trapped electrons by a thermal detrapping process. The ZMR FETs showed evidence of substantial trapping of holes alone in the oxide bulk.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Frequency response theory for multilayer photodiodes

An exact solution is developed for the frequency response of photodiodes composed of multiple spatially uniform layers. Each layer is analyzed separately to obtain a set of linear response coefficients. The response of the multilayer diodes is calculated using matrix algebra. Effects of carrier transit, electron and hole trapping, avalanche decay, and finite absorption length are included in the analysis. The results of R.B. Emmons (1967) and G. Lucovsky et al. (1958) for avalanche photodiodes (APDs) and p-i-n's, respectively, are obtained as special cases. The theory is illustrated by applying it to the separated absorption and multiplication. >

Effects of Arsenic Segregation and Electron Trapping on the Capacitance-Voltage Behavior of Polysilicon and Polycide Gates

AbstractThis paper discusses the effects of dopant segregation and electron trapping on the capacitance-voltage characteristics of arsenic-implanted polysilicon and amorphous Si gate MOS structures fabricated with and without a TiSi2 layer. The effects of gate bias, annealing temperature, silicide formation and polysilicon grain microstructure on the C-V characteristics have also been studied. The results show that insufficient arsenic redistribution at 800°C, coupled with carrier trapping at polysilicon grain boundaries and dopant segregation in TiSi2 causes depletion effects in the polysilicon gate and in turn, an anomalous capacitance-voltage behavior. The depletion tends to increase the “effective” gate oxide thickness and thereby degrade MOS device performance. Higher temperature anneals (≥ 900°C) are sufficient to achieve degenerate doping in the polysilicon gates and avoid the depletion effects.

Toroidal ion temperature gradient driven modes with dissipation and trapped electron effects

An eigenvalue equation is derived for ion temperature gradient driven modes in a toroidal geometry without any expansion in inverse aspect ratio and including the effect of dissipative trapped electrons and electron-ion collisional dissipation. In the electrostatic limit a simple quadratic dispersion relation is obtained. An upshift of the density scale length stability threshold and a downshift of the ηi stability threshold due to electron trapping is demonstrated.

Trapped-electron and generated interface-trap effects in hot-electron-induced MOSFET degradation

A new experimental method is proposed to distinguish the electron-trapping effect in the gate oxide from the interface-trap generation effect in hot-electron-induced nMOSFET degradation. In this method, by selecting the appropriate bias conditions, hot electrons and/ or hot holes are intentionally injected into the oxide region above the channel outside the drain layer, which affects MOSFET characteristics such as threshold voltage and transconductance. The negative charges of electrons trapped in the oxide during hot-electron injection are completely compensated for by the positive charges of subsequently injected and trapped holes, and the trapped electron effect in the degradation is eliminated. Using this method, the causes for hot-electron-induced transconductance degradation (Δg <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> ) are analyzed. As the degradation increases, the trapped-electron effect decreases, and the generated interface-trap effect increases. The relationship of (Δg <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> )_{it}, = <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</tex> (Δg <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> ) -- <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</tex> is obtained, where (Δg <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> /g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> )_{it} is g <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</inf> degradation due to generated interface-traps, and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</tex> are fixed numbers. Furthermore φ_{it}/λ (the ratio of the critical value in hot-electron energy for interface-trap generation to the mean free path of hot electrons in Si) is experimentally obtained to be 5.7 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sup> eV/cm. Using λ = 9.2 nm [1], a value of φ_{it} = 5.2 eV is derived.

Transport properties of thermal oxide films grown on polycrystalline silicon—Modeling and experiments

A theoretical model considering the effects of Fowler-Nordheim tunneling, image-force lowering, first-order trapping kinetics, impact ionization, and asperity-induced field enhancement has been developed to investigate the ramp-voltage-stressed I-V characteristics of the oxide films thermally grown on the polycrystalline silicon. From the ramp-voltage-stressed I-V measurements, the important physical parameters such as average field-enhancement factor, effective total trapping density, trap capture cross section, recombination capture cross section, and dielectric breakdown field can be extracted. Under a ramp voltage stress, it is shown that the serious asperity effect can lead to a larger leakage current and a weaker dielectric breakdown field, but the serious trapping effect may reduce the leakage current and enlarge the dielectric breakdown field. Moreover, dry O <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> oxidation at a higher temperature and steam oxidation at a lower temperature can result in a better quality poly-oxide because the asperity-induced field enhancement is weakened and the electron trapping effect is slightly increased. Besides, high-temperature dry O <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> oxidation can result in a smaller asperity effect as compared with steam oxidation, and the quality of the poly-oxide is deteriorated when the poly-Si substrate is heavily doped because the asperity effect is enhanced.

Influence of a-SiNx: H composition on transfer-doping and electron-trapping effects in a-SiNx: H/a-Si: H superlattices

Abstract The influence of the composition of a-SiNx: H on transfer-doping and electron-trapping effects has been studied in a-SiNx: H/a-Si: H superlattices by measuring in-plane currents. It was found that transfer doping is caused by electron transfer from donor-like states in the a-SiNx: H layer to the a-Si: H layer, and the maximum doping effect occurs at the a-SiNx:H composition of x = 0·85. The origin of the transfer-doping effect is considered to be a structural change in the a-SiNx: H layer. Measurement of photocurrent decay reveals that electron trapping by the a-SiNx: H layer occurs in the superlattice and that it extends to the inner part of the 100 a-thick a-SiNx: H layer. It is also found that the decay time of the photocurrent depends on the a-SiNx: H composition and the temperature. The results suggest that the bottom of the deep-state distribution in the a-SiNx: H layer rises above the Fermi level with increasing x.