Session RP1 - Poster Session VIII.
POSTER session, Thursday afternoon, October 30
Fran Hill Southeast Exhibit Hall, ACC

[RP1.001] Stellerators

[RP1.002] Experimental Plan and Recent Results from HSX

In the HSX QHS configuration, significant reductions in direct loss orbits, parallel viscous damping and neoclassical thermal transport are predicted compared to a conventional stellarator. The quasi-symmetry can be broken through the introduction of a toroidal mirror term in the magnetic field spectrum. The mirror term phased to give the highest trapped-particle fraction at the ECH launch gives poor power absorption and low stored energies; the opposite phase gives absorption and stored energies similar to the QHS case. Experiments on 2nd harmonic ECH at B=0.5 T have been extended to up to 100 kW injected power. Diamagnetic loop measurements of the stored energy are now augmented by Thomson scattering, with Te as high as 700eV observed. Biased electrode experiments have shown reduced damping of flows with symmetry. The DEGAS code has been used to model neutral distribution. Recent results and the future program plan for investigating these configuration differences will be presented.

[RP1.003] Electron Cyclotron Heating at B=0.5T in HSX

Second harmonic X-mode ECH produces and heats the plasma in HSX. Ray tracing calculations predict 40% first pass absorption at a plasma density of 1.5 x 10^18 m-3 and T_e of 400 eV. A set of absolutely calibrated microwave detectors is installed to measure the wave absorption,. The absorption efficiency is very high (about 0.9) in the QHS and Mirror configurations and drops to 0.6 in the Anti-Mirror mode where deeply trapped particles have direct loss orbits. Neutral gas breakdown studies show factors of five higher density growth rates in QHS compared to anti-Mirror. For QHS, the stored energy increases linearly with launched power and plasma density. The radiated power (mainly OII at the edge) increases linearly with absorbed power. Teo measured by Thomson scattering also rises linearly with heating power and reaches ~600 eV at 100 kW of launched power. A four channel ECE system has been implemented on HSX to provide information on the electron temperature profile.

[RP1.004] Characteristics of Electrode Biased Discharges in HSX

The quasi-helically symmetric (QHS) configuration of the HSX stellarator has strongly reduced parallel viscosity compared to conventional stellarators. We use a fast switching biased electrode system to induce plasma flows and a set of Mach probes to measure the flows. Results show that the flow evolution at bias turn on involves two time scales and two directions. The radial electric field is formed on the electrode voltage time scale (».001 msec.) and one component of the flow rises on or near that time scale. A second component of the flow takes longer (».5 msec.) to establish itself. The measured damping times of the bias induced flows are reduced in the QHS configuration by a factor of approximately two compared to configurations with the quasi-symmetry intentionally broken. We model the experiments using neoclassical theory involving parallel viscosity and ion-neutral friction, including a newly developed numerical calculation of the Hamada basis vectors for the 3D HSX geometry. The experimental results agree well with these neoclassical predictions.

[RP1.005] Neutral Density Modeling and Measurements in HSX

The neutral density and particle source rate have been measured experimentally in the Helically Symmetric eXperiment (HSX) using an absolutely calibrated set of 15 Ha detectors. Nine of the Ha detectors make up a poloidal array at one toroidal location, with the other six distributed toroidally around the device. The neutral distribution has been modeled using the Monte Carlo code DEGAS1. 3-D simulations have been normalized to the Ha brightness at the central chord of the poloidal array. The calculations show good agreement in the line-integrated profiles from the Ha arrays when the gas puff is located at the same toroidal plane as the poloidal array. The toroidal Ha array reveals a significant asymmetry in the emission, with the brightness at the gas puff a factor of five greater than the signal on the other side of the torus. Results of simulations of different operating regimes will be presented [1] Heifetz, D.B. et. al., J. Comp. Phys. 46, (1982) 309

[RP1.006] Characteristics of Edge Turbulence in HSX

Langmuir probes have been used to characterize the edge plasma and turbulence in the Helically Symmetric eXperiment (HSX) stellarator. Power spectra of ion saturation current and floating potential (measured on the low field side of HSX) display broad frequency content. Electrostatic particle transport spectra deduced from these signals show a significant contribution for frequencies up to 300 kHz. Total electrostatic particle transport in the edge is approximately 10^16 cm-2s-1 in the outward radial direction, increasing with decreasing radius. Within r/a \sim0.7, a mode at 40-50 kHz is apparent and is responsible for a fraction (<20%) of the particle transport. Phase velocities of the fluctuations are calculated via the two-point correlation technique to be in the ion diamagnetic direction and decrease as the E´B velocities (also in the ion diamagnetic direction) decrease. First electrostatic transport measurements from a probe with access to the high field side of HSX will be presented

[RP1.007] Particle Transport and Density Fluctuations in HSX

Initial estimates of particle transport in the quasi-helically symmetric stellarator, HSX, are carried out using a multichannel interferometer system which has 9 viewing chords with 1.5 cm spacing. Density perturbations are produced by modulating the plasma gas fuelling and the particle source is measured by a multi-channel Ha system. The total radial particle flux is estimated by solving the continuity equation. Diffusion coefficient D and convection velocity V are modeled. Preliminary estimates indicate a diffusion coefficient De\sim1 m2/s. Investigation of fast oscillations evident on the interferometer time-series traces reveals the existence of high-frequency density fluctuations in the range of 25-100 kHz. These fluctuations are noted to have an m=1 nature and are only observed in quasi-helically symmetric plasmas in HSX. Observation of these fluctuations is made during the ECRH pulse when the resonance position is near the plasma core suggesting they may be driven by gradients in the plasma pressure.

[RP1.008] First Thomson Scattering Results on HSX

First results using the Thomson Scattering (TS) System on HSX have been obtained. Based upon the GA divertor TS system design, the HSX TS system consists of several interdependent subsystems; YAG laser, beam transportation, collection optics, fiber optics, polychromators, and data acquisition. It is capable of providing a 10-point radial profile. The central channel is now operational and the rest of the channels will soon be implemented. HSX is currently using 28GHz ECRH at the second harmonic x-mode for plasma production and heating. It is found that the central electron temperature rises linearly with heating power from 30 KW to 100 KW at a fixed density of 1.5\times 10^12cm^-3; in the QHS mode from 300eV to 700eV, and in the Mirror mode from 200eV-600eV. At 40KW heating power, the electron temperature decreases with increasing plasma density, consistent with the diamagnetic measurement of the stored energy. The operation of the system and the detailed results of density scan and power scan for both QHS and Mirror modes will be presented.

[RP1.009] Soft X-Ray Tomography in HSX

Under certain discharge conditions, HSX plasmas exhibit a sudden loss of stored energy followed by fluctuations of the order of few kHz in both the stored energy and the soft x-ray (SXR) signals. These are measured by a diamagnetic loop and a set of PIPS detectors respectively. To help understand the origin of these crashes and the nature of the oscillations, as well as to measure basic plasma properties such as position and shape, a SXR tomography system is under development in HSX. A single array of 20 silicon p-n junction photodiodes is installed on the device and, in the near future, the diagnostic system will be expanded to several arrays in order to obtain tomographic reconstructions of the SXR emission. Initial SXR and stored energy measurements during these crashing discharges as well as the results of the one-array reconstruction will be presented. Implementation details of the complete tomographic system will be shown as well.

[RP1.010] Study of Hard X-ray Emission in HSX

A hard x-ray system employing a CdZnTe detector has been implemented on HSX. The magnetic configuration has been altered between QHS and Mirror mode in order to determine the effect of magnetic ripple on the spectrum and densities of fast electrons. Pulse height analysis of the hard x-ray signals (HX) show the presence of x-ray photons with energy up to 1 MeV. In the QHS mode, the hard x-ray intensity decreases monotonically over two orders of magnitude as the central line averaged density increases from 0.2 to 1.0 x 10^12 cm^-3. In contrast, in the mirror configuration the intensity level is small at the lowest density, peaks at a density of 0.5 x 10^12 cm^-3 and then falls at higher density. In QHS the tail of the spectrum extends to higher energies than for the mirror mode. Increased intensities correlate with the highest stored energies as measured by the diamagnetic loop.

[RP1.011] High ion temperature discharges after Ne glow discharge cleaning in LHD

Ion heating experiment has been extensively carried out in LHD. Using Ne NBI discharges the central ion temperature reached 5keV as a result of the increase in Pi/ni. However, the ion density was dominated by the residual hydrogen. In order to displace the hydrogen to neon, Ne glow discharge cleaning was done. As a result, the ion temperature up to 7keV was obtained in combination of Ar puff. Higher ion temperature was also obtained for C pellet injection. During these discharge a large increment of toroidal rotation was observed. The observed toroidal rotation speed indicated a good correlation with the ion temperature. In the conference, effects of the Ne glow discharge, Ar puff, C impurity pellet and toroidal rotation are discussed.

[RP1.012] Extension of the HINT code and equilibrium computation of LHD with zero rotational transform surface

The three dimensional MHD equilibrium code HINT, which does not assume the existence of nested magnetic surfaces, is modified to extend functions and usabilities. Possibility of a Large Helical Device equilibrium with a zero rotational transform surface is exmined by HINT. The rotational transform profile is assumed to be controlled by amount of a net toroidal current. In the equilibrium, we find existence of the equilibrium but formation of two n=0 islands composing the homoclinic-type structure near the center, where n is a toroidal mode number. As further increase of beta, three n=0 island structure is formed, while the homoclinic-type nature is retained. A maximum in the Shafranov shift is observed as the increase in the amount of the net current.

[RP1.013] Eigen surface cyclotron modes of the stellarator plasma column

The eigen surface modes of magnetically confined plasma column can be excited during the ion cyclotron resonance heating (ICRH) in stellarators plasma. These modes can be localized in a narrow region at the plasma edge. Therefore the essential part of the ICRH power will be absorbed at the plasma edge without the effective plasma heating in the center. Due to this reason the ICRH system has to avoid operating in the range of the parameters which provide a high probability of the surface mode excitation. As an example, the eigen surface modes of LHD have been studied to estimate the penetration depth of wave field into plasma column and calculate their radial distribution. The calculations have been carried out for the nonuniform radial density and temperature profiles. The eigen wavelength of the surface perturbations at harmonics of the cyclotron frequency are reported. The dependence of the eigen wavelength on the plasma density gradient has been studied.

[RP1.014] Development of Heavy Ion Beam Probe System in CHS

Heavy Ion Beam Probe (HIBP) is a very powerful tool that is able to simultaneously detect potential, density, magnetic field and their fluctuations in interior of high temperature plasmas. In CHS, so far, the HIBP has revealed interesting behavior of the plasma interior, such as bifurcation, internal transport barrier, MHD phenomena, by observing potential profile and density fluctuation. Now we are making efforts to make a full use of the other abilities of HIBP, that is, to detect potential fluctuation and magnetic field fluctuation. In order to study potential fluctuation, a new ion source has been developed to increase beam current density. With the new ion source, we have obtained current 3~5 times larger than the previous one. This improvement allows us to detect the potential fluctuation, and also density fluctuation with higher S/N ratio. As for the magnetic field fluctuation estimated from the beam displacement, the interpretation of the signal is quite difficult because of 3D magnetic field configuration of CHS. In the first place, a numerical calculation has been performed in axisymmetric toroidal field i. e., tokamak configuration, in order to obtain a basic perspective of the relationship between the beam displacement and local magnetic field fluctuation. In our presentation, we will describe the resultant progress of HIBP measurements, and the recent results of density and potential fluctuation measurements in the CHS stellarater.

[RP1.015] Progress in Analysis and Construction of NCSX.

The National Compact Stellarator Experiment (NCSX) is being developed to explore the high-\beta stability and confinement of a quasi-axisymmetric stellarator with average aspect ratio 4.4. It is designed to be passively stable to the ballooning, kink, vertical, Mercier, and neoclassical tearing modes for \beta > 4% without the need for external feedback or conducting walls. The bootstrap current generates only 1/4 of the magnetic rotational transform at \beta = 4% (the rest is from the coils). As a consequence, stability properties are relatively insensitive to the pressure profile shape. The quasi-axisymmetric magnetic field produces thermal neoclassical transport similar to equivalent tokamaks and adequate fast-ion confinement. It also produces low damping of flows in the toroidal direction. The consequence of the low toroidal flow damping on the expected turbulence will be discussed. The coils have been designed to provide good flux surfaces and to flexibly investigate the predicted MHD stability and transport properties. Prototypes of the coils and vacuum vessel are under construction. Fabrication of the production coils and vessel is scheduled to start next year. Development of a national team for research and diagnostic development will also begin next year.

[RP1.016] Progress in Technology Research and Development for NCSX

The physics mission of the National Compact Stellarator Experiment imposes demanding requirements on the manufacturing technologies for the modular coils and vacuum vessel. These components must satisfy complex three-dimensional shape specifications to high accuracy while meeting their performance requirements. A research and development program is being carried out by PPPL, ORNL, and industry to demonstrate the necessary technologies. Advanced numerical modeling, metrology, and manufacturing methods are being developed to optimize the fabrication of the modular coil winding forms (by casting and machining) and vacuum vessel (by forming and welding). Coil winding developments include flexible cable design, epoxy impregnation of complex-shaped windings, metrology, composite materials properties, and dimensional control during the winding process.

[RP1.017] Two-Fluid Steady States of High Beta Stellarators

Stellarator steady states and stability obtained from two-fluid nonlinear relaxation are very different from MHD predictions and may allow higher beta. Numerical simulations of the NCSX high beta stellarator with the M3D two-fluid code show that a strong equilibrium pressure gradient near the plasma edge (``H-mode'') creates a region where diamagnetic drifts and related effects drive a steady state poloidal flow on fast MHD-like time scales, well before the neoclassical parallel viscous stress acts to reduce v_i\theta. The flow in the strong \nabla p region is similar over a large range of p_e/p_i, in the ømega_*i-direction with magnitude significantly smaller than an average diamagnetic velocity based on (p_e+p_i)/2. The flow robustly stabilizes high mode number resistive ballooning/interchange modes that limit MHD stability in simulation, but generally not in experiment. In addition, electron two-fluid effects can increase magnetic island growth rates on interior rational surfaces. These effects increase with \beta and p_e/p_i, suggesting that the best stellarator confinement may exist in hot-ion operation. The beta limit may be a gradual one due to the overlap of large, although contained, magnetic islands that destroy the effective plasma confinement, rather than catastrophic instability.

[RP1.018] Neoclassical transport in NCSX

Neoclassical viscosities and cross-field particle and heat fluxes are presented for projected NCSX conditions. Calculations based on full 3-D magnetic configurations are compared with predictions for equivalent axisymmetric configurations and an analytic model for neoclassical helical transport. The transport matrix for 3-D configurations is based on monoenergetic diffusivities, obtained here from the DKES code, and from MOCA, a Monte Carlo orbit code. The pitch-angle-scattering collision operator does not conserve momentum, so we use a procedure proposed by Sugama to obtain the corrected transport parameters. The magnetic geometry of NCSX can be varied by changing the internal bootstrap currents and the independently driven currents in the external modular field coils. To obtain efficient estimates of the neoclassical transport for each new configuration, the monoenergetic diffusivities are represented with physically motivated basis functions. We also compare to predictions of the ambipolar radial electric field and the neoclassical helical particle and energy fluxes given by the analytic Shaing-Houlberg model (with an improved treatment of the \nu regime).

[RP1.019] Eliminating islands in high-pressure free-boundary stellarator magnetohydrodynamic equilibrium solutions.

Magnetic islands in free-boundary stellarator equilibria are suppressed using a procedure that iterates the plasma equilibrium equations and, at each iteration, adjusts the coil geometry to cancel resonant fields produced by the plasma. The coils are constrained to satisfy certain measures of engineering acceptability and the plasma is constrained to ensure kink stability. As the iterations continue, the coil geometry and the plasma simultaneously converge to an equilibrium in which the island content is negligible. The method is applied with success to a candidate plasma and coil design for the National Compact Stellarator eXperiment [Phys. Plas., 8(5),2083 2001].

[RP1.020] Heat Load on the NCSX First Wall

We have used magnetic field data generated by the PIES 3D MHD equilibrium code(M50 coil set) and a new vacuum field code(Michael Drevlak Max Planck Institute for Plasma Physics, Greifswald, Germany, private communication), together with the latest numerical model of the first wall (Art Brooks, PPPL, private communication) to compute wall heat loading in the National Compact Stellarator Experiment (NCSX) as a function of position. Field lines originating just outside the last closed magnetic surface are followed until they intersect the wall by integrating the field-line equations of motion in cylindrical coordinates using the LSODE integrator. Magnetic field values are interpolated with arbitrary-order splines. The local (nonuniform) heat flux is estimated from the density and incidence angle of escaping field lines.

[RP1.021] Overview of the QPS Experiment.

The Quasi-Poloidal Stellarator (QPS) is a very-low-aspect-ratio (R/a = 2.7) compact stellarator in which the dominant magnetic field components are poloidally symmetric in flux coordinates, which leads to large reductions in: neoclassical transport at low collisionality; bootstrap current; and poloidal viscosity, which allows large E x B poloidal flows for suppression of anomalous transport. The magnetic configuration is relatively insensitive to increasing beta. The experiment under design has R = 0.95 m, a = 0.35 m, B = 1 T for a 1.5-s pulse, and P(heating) = 2-4 MW. Nine independent coil currents allow varying: neoclassical transport by a factor \sim25, degree of poloidal symmetry by a factor \sim10, and poloidal viscosity by a factor of 530. QPS can study regimes in which either the anomalous transport or neoclassical transport is dominant, ballooning stability limits at beta = 2.5%, and equilibrium robustness at finite beta. Recent progress on physics issues, the relationship to other stellarator concepts, the QPS project status, and the proposed experimental program are presented.

[RP1.022] Design of the QPS Experiment.

The Quasi-Poloidal Stellarator, QPS, is a low-aspect-ratio (R/a = 2.7), compact stellarator under design at ORNL. The device parameters are R = 0.95 m, a = 0.35 m, and B = 1 T for 1.5 s with 2 MW of ECH and 3.5 MW of ICRF for plasma heating. A nonplanar modular coil set provides the primary magnetic field configuration. It has two field periods with ten modular coils per period. Due to stellarator symmetry, there are only five different coil types. Flexible copper cable conductor will be wound on a stainless steel form, vacuum impregnated with epoxy, and canned for vacuum compatibility. The coil form allows the coils to be connected into an integral structural shell. Unlike most stellarators, a vacuum vessel surrounds the coils rather than fitting inside the coils, allowing excellent access for plasma diagnostics and heating. In addition there are external vertical field and toroidal field coils and an ohmic current solenoid for configuration flexibility. First plasma operation is planned for the end of 2007. Details of the engineering design and analysis will be presented.

[RP1.023] Confinement, Flow Damping and Flexibility of Quasi-Poloidal Stellarators

Quasi-poloidal (QP) stellarators have achieved levels of optimization that significantly suppress neoclassical transport relative to anomalous levels. QP symmetry also allows poloidal flow damping to be less than toroidal flow damping. This ordering (reversed from that of tokamaks) should allow more efficient control of the radial electric field with less required momentum input. An analysis of flow dynamics in such systems has been initiated using viscosity coefficients that can be derived from the usual transport coefficients (density/energy diffusion, bootstrap current. resistivity enhancement). The dependencies of these coefficients on plasma parameters such as collisionality and electric field will be analyzed and implications for flow evolution discussed. By varying the modular and vertical field coil currents in proposed QP systems, substantial flexibility has been demonstrated theoretically with respect to neoclassical confinement, quasi-poloidal symmetry, flow damping, and magnetic island widths. Variations of a factor of 20 in low collisionality neoclassical transport rates and a factor of 5 in quasi poloidal symmetry can be achieved. Work supported by U.S. Department of Energy under Contract DE-AC05 00OR22725 with UT-Battelle, LLC.

[RP1.024] Impact Of Plasma Current Profile And External Coil Currents On Ideal Mhd Stability In The Quasi-Poloidal Stellarator

This work examines the impact of both the plasma current profile and external coil currents on ideal MHD stability in the Quasi-Poloidal Stellarator (QPS) [1]. The reference coil set for QPS contains 20 modular coils with 5 different coil shapes with each of the 4 coils of a given shape forming a group with a separate power supply. In addition to the modular coils, there are also a set of toroidal field coils and 3 pairs of vertical field coils. This coil set allows for flexible variation of the magnetic configuration. QPS has been designed to run with a bootstrap aligned current profile but will also have the capability of driving Ohmic current. By varying both the external coil currents and the plasma current profile, the ideal MHD stability properties of QPS can be greatly impacted. The infinite-n ballooning \beta limit can vary by a factor of ~2. The marginal ballooning \beta limit is lowest with an Ohmic current profile, highest with a bootstrap current profile, and can be varied between these limits with a combination of Ohmic and bootstrap current. As expected, finite-n ballooning modes (for n up to 19) have higher \beta limits, roughly 50-100% higher than the infinite-n limit for QPS. Kink and vertical modes are stable in QPS to much higher plasma \beta (> 5%). [1] J. F. Lyon, et al., “Physics and Engineering Design of a Very-Low-Aspect-Ratio Quasi-Poloidal Stellarator”, in preparation for Nucl. Fusion, (2003).

[RP1.025] A Breeder Algorithm for Stellarator Optimization

An optimization algorithm that combines the global parameter space search properties of a genetic algorithm (GA) with the local parameter search properties of a Levenberg-Marquardt (LM) algorithm is described. Optimization algorithms used in the design of stellarator configurations are often classified as either global (such as GA and differential evolution algorithm) or local (such as LM). While nonlinear least-squares methods such as LM are effective at minimizing a cost-function based on desirable plasma properties such as quasi-symmetry and ballooning stability, whether or not this is a local or global minimum is unknown. The advantage of evolutionary algorithms such as GA is that they search a wider range of parameter space and are not susceptible to getting stuck in a local minimum of the cost function. Their disadvantage is that in some cases the evolutionary algorithms are ineffective at finding a minimum state. Here, we describe the initial development of the Breeder Algorithm (BA). BA consists of a genetic algorithm outer loop with an inner loop in which each generation is refined using a LM step. Initial results for a quasi-poloidal stellarator optimization will be presented, along with a comparison to existing optimization algorithms.

[RP1.026] Development of a Prototype Three-Dimensional Equilibrium Reconstruction Code

An initial step toward developing an efficient equilibrium reconstruction code for three-dimensional (3D) plasmas has been taken. The 3D magnetics diagnostics codes V3RFUN and V3POST can be used to rapidly compute the expected magnetic signals in external loops arising from both external coil currents and internal stellarator plasma currents. The V3RFUN code is executed once initially (for a prescribed set of stellarator coils) to generate a database of diagnostic responses valid for arbitrary plasma currents. Then, the V3POST code is executed repeatedly from within the 3D optimization code STELLOPT to compute diagnostic signals arising from the MHD pressure and iota profiles. The optimizer determines the profiles which best match the measured external diagnostic signals. The relative computational efficiency of the V3POST evaluations renders this procedure numerically feasible and will provide a prototype for the eventual development of an optimized 3D reconstruction code (V3FIT). The prototype code will be used to examine numerical features of the reconstruction process - such as the level of convergence needed to estimate the signal gradients - which will guide the development of the optimized V3FIT code.

[RP1.027] Reconstruction of 3D Equilibria from Flux Surface Knowledge Only

Using the properties of the Grad-Shafranov (GS) equation, Christiansen and Taylor (CT) have shown(J.P. Christiansen, J.B. Taylor, Nucl.Fusion) 22, 111 (1982). that complete MHD equilibria may be obtained for axisymmetric tokamaks with noncircular cross-sections, provided that one initially knows only the shapes of the flux surfaces. Starting from a 3D generalization of the GS equation(L.M. Degtyarev, V.V. Drozdov, M.I. Mikhailov, V.D. Pustovitov, V.D. Shafranov, Sov. J. Plasma Phys.) 11 22 (1985), we have recently demonstrated(H.E. Mynick, N. Pomphrey, Phys. Plasmas) 9,1050 (2002) that this remarkable result can be extended to 3D systems like stellarators.

A code to practically implement this earlier formal result is nearing completion. For testing, the code takes plasma shapes from VMEC equilibria, and the predicted profiles are compared with those from VMEC. With it, we will test if this extended CT method may be used as a practical diagnostic to use emissivity data from stellarators, and what the sensitivity of the results are to uncertainties and sparseness of the experimental data.

[RP1.028] Construction Progress of the Compact Toroidal Hybrid

The Compact Toroidal Hybrid (CTH) is under construction at Auburn University. CTH is a stellarator/tokamak hybrid device that will use an ohmic current of I_p\leq 50kA to investigate both ideal and resistive current-driven instabilities in low aspect ratio stellarators, and to carry out 3-D plasma equilibrium reconstruction tests. The edge vacuum rotational transform in CTH is variable from t_V(a) = 0.2 to 0.5 with a current-generated transform t_J(a) \leq 0.5. The average plasma major radius is R_o = .75m, the vacuum vessel minor radius is a_v = 0.29m and the maximum average plasma minor radius is a_p = 0.20m. CTH is a low \beta (\leq0.005) machine and will operate at magnetic fields B_o \leq 0.5T. Target plasmas for ohmic current stability studies will be generated by an 18GHz klystron operating at ømega=2ømega_ce. The CTH vacuum vessel has been cleaned, and outfitted with heater pads and magnetic diagnostics. The poloidal field coils have been wound and helical field coil winding will begin shortly after delivery of the helical coil frame. The CTH device is scheduled to be completed and operational in mid-2004. Supported by US DoE Grant DE-FG02-00ER54610

[RP1.029] High Beta Helical Equilibrium

An analytic solution of the helical Grad-Shafranov equation subject to the constraints \beta=2\mu_0p/B^2\sim O(1) and q=d\Phi_T/d\Phi_P\sim O(1) is presented. The pressure profile p(\psi) and the safety factor profile q(\psi) are specified. The solution takes the form of a boundary layer problem with the inverse aspect ratio \epsilon=a/r_0 as an expansion parameter. In the core region, \psi is a function of only the major radius so the flux surfaces are vertical lines. The boundary layer is located at the wall and the Shafranov shift is of the order the minor radius. A solution with vanishing toroidal current is also presented. This condition imposes a constraint relating p(\psi) and q(\psi) such that only one profile may be independently specified. DOE Contract Number: DOE-AC02-76-CHO3073.

[RP1.030] The effect non-linear wave-particle interaction on ECRH with helical ripple stellarators

Particles trapped in the helical ripples in stellarators bounce back and forth along the magnetic field line. Trapped particles get energy from the wave repeatedly. When the turning points are far from the resonance region, the heating process is stochastic and the wave-particle interaction makes a small perturbation on the distribution function. The rate of velocity-diffusion for the leading order distribution function is proportional to the sum of squares of the wave amplitudes. This is the, so called, "quasi-linear diffusion theory" since linear absorption and quadratic diffusion are used. However, quasi-linear theory is not valid when the turning points are very close to the resonance region and the particles stay in the resonance region such as in second harmonic X mode ECRH. When this is the case, the wave and particles are correlated and the stochastic process is not viable. To obtain ECRH heating in this nonlinear regime, collisions need to be explicitly accounted for. Nonlinear wave-particle interaction for deeply trapped particles has been described in previous work. In this work, the equivalent diffusion term when nonlinear wave-particle interaction dominates will be derived. It is desired to look for steady-state solution of the Fokker-Planck equation when the heating term and the Coulomb collision term are balanced in order to analyze power deposition.

[RP1.031] JET

[RP1.032] Real-time profile control in JET for steady state advanced tokamak operation

Real-time control of the plasma profiles (current, pressure and flows) is a key issue to sustain steady state discharges with internal transport barriers (ITB) and a large bootstrap current fraction. In order to simultaneously control the current and pressure profiles in JET ITB discharges, a multi-variable model-based technique has been proposed. It is based on a truncated singular value decomposition of an integral model operator and retains the distributed nature of the plasma parameter profiles. The related algorithms have been implemented in the JET control system, and applied to the control of the current profile in reversed shear plasmas using three actuators (neutral beam injection, ion cyclotron heating and lower hybrid current drive). Successful control of the q-profile has been achieved in quasi steady state conditions with a significant fraction of bootstrap current. In these experiments the strength of the ITB's was marginal during the control phase, due to the chosen q-profile setpoints and to the moderate heating power which was requested by the controller. Hence, further experiments aiming at the simultaneous control of the current profile and of the normalized ITB temperature gradient are being pursued and first results will be reported.

[RP1.033] Edge Transport Barrier Behaviour in JET ELMy H-mode Plasmas

The ELMy H-mode plasma configuration is the most likely candidate, at the moment, for ITER and fusion based power plants. Hence it is very import to understand the workings of these plasmas. Recently on JET a large database of edge transport barrier, ETB, data has been gathered using the edge LIDAR Thomson scattering Diagnostic. This had been impossible before, due to system resolution limitations. Now a plasma configuration has been designed (DOC-U) with the last closed fluxsurface tangential to the edge LIDAR laser line of sight, effectively increasing the spatial resolution to ~ 2 cm. The results obtained from this database are shown [1] to be contradictory to expectations and other experiments [2,3]. It shows that the ETB gradients decrease with an increasing Greenwald density fraction and the pedestal width increases. In this paper these results are expanded and further analysed, to achieve a better understanding of the ELMy H-mode plasma edge behaviour in JET.

[1] M. Kempenaars, et al, 30th EPS 2003, St. Petersburg, 5-11 July 2003 [2] G. Saibene, et al, Plasma Physics and Controlled Fusion 2002 [3] A. Kallenbach, et al, 30th EPS 2003, St. Petersburg, 5-11 July 2003

[RP1.034] Modeling effects of local surface properties on heat flux deposition in the JET divertor

Understanding heat flux deposition from ELMs is an essential issue for a next step fusion device. A high time resolution infrared system is used in JET to measure the surface temperature distribution and its evolution on the divertor target plates. Previously, an empirical technique was developed, based on a flexible 1D model calculation, to assess possible complications due to surface layer properties, such as poorly adhered a-C:D layers [1]. The model validation used data from JET DOC-L discharges (DOC-L: inner and outer strike points positioned for optimized infrared measurements) with programmed constant L-mode power steps. The effect of layers was identified for the inner tile surface. In this paper we compare the 1D model for surface temperature evolution with results of 3-D modeling with the CASTEM-2000 thermal code, for these DOC-L power step cases. The 1-D values are shown to approach the 3-D results as the model power deposition width increases, showing that there is a absolute 30% accuracy for the 1D model along with a well-supported validation for its use in scaling studies. Additional modeling describing the role of layers and also of small localized heat sinks (dust), as is suggested for similar cases [2], will be presented. [1] Y. Corre et al, EPS 2003, St. Peterburg [2] E. Delchambre et al, J Nucl Mater 2003

[RP1.035] Deuterium to Helium Plasma-Wall Change-over Experiments in JET with Reversed Toroidal Field

The deuterium and helium dynamics in the plasma and subdivertor regions of JET are compared during a sequence of similar ohmic and NBI pulses where 100% He gas is injected into the JET vacuum vessel, whose graphite walls were previously saturated with deuterium. After the first six He fueled change-over discharges, only He plasma operation was performed. The He concentration is measured in the sub-divertor with a species selective Penning gauge. Comparison of the time dependence of the divertor concentrations with those at the edge and strike point shows significant differences during the first six discharges. Data has been obtained for both the normal and reversed toroidal magnetic field direction. These differences along with a global He particle balance is used to assess the status of the wall saturation over the initial 6 - 7 He change-over discharges.

[RP1.036] Resistive Wall Mode studies on JET

It is important to determine the scaling of the stabilising requirements for RWMs to extrapolate to ITER and beyond. The underlying damping determining RWM stability has been probed with applied error fields on JET; similar to the methods used on DIII-D. Experimental results show good agreement with modelling with the MARS code, in particular on the amplification of the applied error field as beta increases towards the ideal limit. Analysis of old data has also revealed instabilities with many RWM characteristics, possibly giving another route for cross machine comparison. (Funded by EURATOM under EFDA)

[RP1.037] Core and Edge MHD Studies in JET Reversed B Experiments

JET operation with reverse B changes the NBI heating to counter injection, while the normal field has co-NBI. One consequence is to change NBI induced rotation and its effect upon MHD has been studied. Shorter sawtooth periods were obtained with counter-NBI. With co-NBI, the sawtooth period increased with power, while with counter-NBI, the sawtooth period reduced to a minimum at 4 MW. A candidate mechanism consistent with the salient features of this trend is the dependence of kinetic internal kink stabilization on sheared toroidal rotation and central energetic ion penetration in reverse B plasmas. ELMs and Washboard modes in high density ELMy H-mode plasmas, as well as edge MHD modes in the quiescent H-mode (QHM) regime were studied. Although the QHM was not obtained, an edge continuous n=1 mode (with several harmonics) was observed during ELM-free phases. The edge MHD features were similar to the "edge harmonic oscillation" of DIII-D and also similar to earlier JET counter-NBI experiments where the n=1 external kink (outer mode) was found to be more unstable than with co-NBI.

[RP1.038] Size Scaling on Carbon Screening by a Tokamak Divertor

Plasma impurity content depends upon impurity sources, fuelling efficiency, and confinement. In JET, carbon is the primary impurity, and its fuelling efficiency has been studied using methane gas injection. In this poster, EDGE2D modelling of the JET experiments and similar AUG experiments are extended to ITER. For carbon injected at the mid-plane, screening depends upon: the SOL temperature, the carbon ionisation depth, the thermal force near the divertor entrance, the parallel distance to the divertor, and the SOL flux expansion. For carbon injected at the strike points, the carbon screening depends upon the carbon ionised in the region where the thermal force is strong. In JET, carbon, originating inside the divertor, had typically 5% chance of reaching the main chamber SOL, but from the main chamber SOL had about 10% chance of contaminating the core plasma. By contrast, carbon originating in the main chamber, was typically 95% ionised inside the main chamber SOL but had only 4 to 10% chance of contaminating the core. Since the ITER SOL will be hotter, and the ITER divertor will be larger than JET, the calculated carbon screening is much better.

[RP1.039] Testing of the JET ITER-Like ICRH Antenna High Power Prototype

A High Power Prototype (HPP) of the JET ITER-like ion cyclotron resonance heating (ICRH) antenna has been constructed and tested in a joint collaboration between Oak Ridge National Laboratory, Princeton Plasma Physics Laboratory, and EFDA-JET/UKAEA. The HPP consists of one quadrant of the full JET-EP ICRH antenna. Internal matching capacitors are utilized in a circuit that maintains a voltage standing wave ratio at the input < 1.5 over a factor of ten range in resistive loading. The HPP is being tested in vacuum at full capacitor design voltage and current for up to 10-s pulses. The HPP design and status of the test results will be reviewed. Measurements of the currents, voltages, input impedance, and temperatures will be discussed including low power measurements of the tuning range and sensitivity to capacitor position. Measurements will be compared to circuit analysis and modeling codes. The impact of HPP results on the design of the final JET ITER-like ICRH antenna will be discussed.

[RP1.040] Simulations and Stability Analysis of JET H-mode Triangularity and Power Scans

Four JET H-mode discharges in a triangularity scan and three JET H-mode discharges in a power scan are simulated using the JETTO code. The simulation results are analyzed using the HELENA and MISHKA stability codes. In the JETTO simulations, periodic ELM crashes are triggered either by a pressure-driven mode or by a current-driven mode. For the triangularity scan, the variation of the pedestal height, which is caused by the variation in MHD stability of finite-n ballooning/peeling modes with triangularity, is consistent with experiment. For the power scan, it is found that ELMs are triggered mainly by current-driven modes, where a significant part of the edge current is the bootstrap current that is driven by the edge pressure gradient. As a result of inductive effects between ELM crashes, higher-pressure gradients are achieved in the pedestal, as the heating power is increased, consistent with experiment.

[RP1.041] Ignitor, Fire, and ITER

[RP1.042] The Ignitor Experiment: Fusion Burning Phase, Isotopic Control and Optimal Injected Heating

Ignitor^1 has been the first experiment proposed and designed to reach ignition conditions, and remains unique in preserving this goal. At ignition where plasma heating due to the fusion reaction products compensates for all forms of energy loss, the thermonuclear instability sets in and an appropriate set of two nonlinear equations that include the tendency of the plasma temperature to explode^2 and the quenching due to pressure gradient driven modes, is introduced to describe the resulting (oscillatory) burning phase. Quasi-stationary subignited regimes are identified using the JETTO transport code by controlling the isotopic composition of the plasma and applying appropriate pulses of ICRH heating to the plasma column. When the DT mixture is optimal (50-50), the ICRH power needed to shorten significantly the approach to ignition can be as low as 3MW if applied at the time when the ideal ignition condition is reached.

^1B. Coppi et al., Nucl. Fusion 41, 1253 (2001) ^2A.C. Coppi and B. Coppi, Phys. Plasmas 6, 1470 (1999)

[RP1.043] Optimal Temporal Evolution of the Magnet Currents of the Ignitor Machine

The Ignitor Magnet System is comprised of 24 Toroidal Field Coils (TFC) and 12+2 pairs of Poloidal Field Coils (PFC), including the Central Solenoid and the EM Press. Currents in the TFC and each of the PFC are regulated by dedicated thyristor converter units. Starting from a first possible scenario for the magnet currents and voltages evolution, an iterative optimization process of the total installed electrical power has been performed, by taking into account several elements: smoothing of the current evolution, reduction of the voltage step variations by switching on proper resistors, modifying the coil turns and the coil current consequently, connecting in series coils with similar current behaviour, eliminating coils if poorly effective on the plasma, using converter schemes (sequential control or internal feedback control) suitable to reduce the reactive power requirements and ac current harmonics. In this way the total installed power in the Ignitor Power Supply System is about 2200 MVA instead of about 3400 MVA calculated before the optimization process. In addition the power requirements (max active power, max reactive power, max active power negative/positive steps) result compatible to the 400 kV Grid operational requirements.

[RP1.044] Load Assembly of the Ignitor Machine with 3D Interactive Virtual Reality

The main purpose of this work is to assist the Ignitor team in every phase of the project using the new Virtual Reality Technology (VR). Through the VR it is possible to see, plan and test the machine assembly sequence and the total layout. We are also planning to simulate in VR the remote handling systems. The complexity of the system requires a large and powerful graphical device. The ENEA?s "Advanced Visualization Technology" team has implemented a repository file data structure integrated with the CATIA drawing cams from the designer of Ignitor. The 3D virtual mockup software is used to view and analyze all objects that compose the mockup and also to analyze the correct assembly sequences. The ENEA?s 3D immersive system and software are fully integrated in the ENEA?s supercomputing GRID infrastructure. At any time all members of the Ignitor Project can view the status of the mockup in 3D (draft and/or final objects) through the net. During the conference examples of the assembly sequence and load assembly structure will be presented.

[RP1.045] The Fast Pellet Injector Program for Ignitor

The characteristics of a fast pellet injector for the Ignitor ignition experiment have been identified. In order to produce sufficiently peaked density profiles during the initial phase of the current ramp and to sustain them along the flat top phase, a multiple injector capable of shooting pellets of variable sizes will be developed. The program involves the collaboration of the ENEA Laboratory at Frascati and the Fusion Technology Group of Oak Ridge. The initial activities will be devoted to the development of a 4 barrel, double stage gun able to reach speeds up to 4 km/s. The compact size of the Ignitor machine makes injection from the high field side unpractical, while it is unclear that a vertical injection close to the magnetic axis will be beneficial. Simulations performed with the PELLET code(W.A. Houlberg, et al., Nuclear Fusion) 28, 595 (1988), on the other hand, indicate that pellet speeds of 3-4 km/s would allow a sufficient particle penetration from the low field side, particularly during the initial current ramp up phase, when the plasma temperature is still relatively low and good control of the density profile is most desirable.

[RP1.046] Disruption Scenarios and their Effects on the Ignitor First Wall

The 2D simulation of the reference plasma disruption in Ignitor (a fast current quench following a Vertical Displacement Event - VDE), has been performed using the MAXFEA code. The resulting excitation loads have been used in a detailed 3D model to evaluate the electromagnetic loads in the first wall and its supports. The model of the most loaded region of the first wall has been performed using a zooming procedure that allows the replacement of the out-of-model plasma, poloidal coils and passive structures with current filaments surrounding the modeled region. To identify the most stressed tiles under the effect of the electromagnetic loads due to the VDE, a preliminary structural analysis has been carried out using a simplified model. Then a more detailed model has been performed for these tiles, with the aim of evaluating the stress on the screw and the pre-load needed to avoid detachment between the tile and the tile-carrier. The thermal loads on the tiles are evaluated independently and the implications of possible other disruption scenarios are discussed.

[RP1.047] First Wall of the Ignitor Machine

The design of the first wall of the Ignitor machine^2 involves a system of Molybdenum tiles that constitute a “continuous limiter” surrounding nearly all of the plasma column. Sets of tiles are connected to tile carrying plates that can be replaced by an appropriate remote handling system. The adopted solutions for the attachment of each tile to the tile carrier and of the tile carrier to the plasma chamber on the inboard side and to the appropriate supporting structure on the outboard side are described. The relevant structural analysis is presented.

^2B. Coppi et al., Nucl. Fusion 41, 1253 (2001)

[RP1.048] Advanced Design and Construction of Toroidal Magnet Plates

The main elements of the toroidal magnet Ignitor^2 are normal conducting plates each carrying up to 357.5 kA. According to the machine design these plates are to be cooled at 30 K that is the optimal temperature for the material that has been chosen (Copper OFHC) Given the properties of this material a new solution for the cooling of each plate has been adopted. The construction of two prototypical full size plates is underway taking into account the experience gained in the previous construction of 10 plates that were assembled into a full toroidal field coil. The toroidal magnet of Ignitor is made of 24 coils.

^2B.Coppi, A.Airoldi, F.Bombarda et al., Nucl. Fusion 41,1253 (2001).

[RP1.049] Advances in the Structural Design of the ICRH Antenna for the Ignitor Experiment

The Faraday shield (FS) and the strap array are the in-vessel elements of Ignitor’s ICRH antennas. Detailed non-linear and 3-D finite element analyses were performed to analyze their thermal and structural behavior. Dynamic disruption loads are superposed to quasi-static thermal loads relying on the different time scale of the two phenomena. Disruption loads on the FS are computed by using an inductive-resistive lumped parameter circuit and those on the straps by using a resistive circuit. Disruption loads on the Faraday shield bars are minimized by electrically insulating one side of the bars. The effects of induced eddy currents on the straps are taken into account. Moreover, the latest simulations model the thermal loads over the straps taking into account the presence of the FS. The mechanical design of each component is optimized in order to facilitate the full remote handling installation, to minimize the stresses induced during plasma pulses and to guarantee a high degree of reliability. Finite element analyses show that the FS and the straps will be able to withstand the expected loads.

[RP1.050] Analysis of the Connection of Ignitor to the Rondissone Node of the European Electrical Grid

The basic requirements for the power supply of the Ignitor machine and the effects of its connection to the electric node of Rondissone, in the North-West of Italy, are described. This study identifies the possible technical restrictions associated with a flexible and meaningful operation of the Ignitor experiment. The study was carried out by GRTN, the Italian Independent power System Operator (ISO), taking into account two typical extreme conditions of the grid, at the highest (winter) load and at the lowest (summer) load. The worst scenario generates peak power consumption of about 1095 MW and 860 Mvar, as well as a peak power delivery of 428 MW. Results show that when the Ignitor machine operates at its normal conditions it doesn’t generate disturbances on the system frequency that exceed the allowable range +/- 0.2% (50%) prescribed by the GRTN grid code. The generated voltage interferences at the substation of Rondissone are acceptable. Local systems of reactive power compensation (SVC) are envisaged to minimize voltage disturbances.

[RP1.051] Design Criteria for the Columbus Experiment

In principle, it is possible to scale up the parameters of Ignitor, the only experiment proposed and designed to reach ignition, but within a narrow range preserving the ability to maintain this goal with reasonable margins against the onset of macroscopic instabilities, given the high plasma pressures required. A strong ohmic heating is to be maintained so that ignition can be reached even in the case of a failure of the ICRH system. The Columbus experiment^2 is proposed as a parallel US project to the Ignitor program carried out in Italy and is geometrically self similar to this, the dimensions being increased by 25/22 (R_o=1.50 m) and the volume by about 50%. The most important parameter design guideline is the value of the mean poloidal field (\sim3.4T ). The plasma current I_p\sim12.2MA is close to that of the ITER-Feat concept for the same value of the safety factor q_95(\Psi)\sim3.6. As in the case of Ignitor and ITER-Feat, the plasma current redistribution time and the duration time of the plasma burning state are comparable. Columbus incorporates all the technological solutions developed for Ignitor.

^2B. Coppi and M.F. Salvetti, MIT (RLE) Report PTP 02/06 (December 2002, Cambridge, MA)

[RP1.052] Preliminary Thermal Analysis of the Toroidal Field Coils in the Columbus Machine

The Columbus experiment^1 is geometrically self similar to Ignitor, the linear dimensions being increased, roughly, by the factor 25/22 (R_o=1.50 m) and the volume by about 50%. Characterized by decreased current densities in the toroidal and poloidal magnet systems, Columbus allows for longer plasma pulses relative to Ignitor. We identified a reference current pulse in the toroidal magnet that has a flattop extended to about 7.6s (disregarding the contribution of high energy neutrons). The Toroidal Field System is made of 24 OFHC copper coils. Similarly to Ignitor, the highest current flowing in each of the coil plates is 357.3 kA (the total magnet current being I_M=94.38 MA-turn), and the acceptable temperature in the magnet ranges from 30 K to about 240 K. We carried out coupled electromagnetic and thermal 3-D finite element analyses to determine the temperature distribution within the magnet and we compared our results to those obtained by using the 2-D FORTE code. The magneto resistive effect is presently being included in our model.

^1 B.Coppi and M.F.Salvetti, MIT (RLE) Report PTP 02/06 (December 2002, Cambridge, MA).

[RP1.053] Continuity of Confinement Regimes in High Field Experiments

High magnetic field experiments have shown consistently that under ohmic heating conditions, the energy confinement time tends to increase with density as long as its profile is significantly peaked. This peaking can occur spontaneously as for the Alcator A and FT experiments or be produced by injection of pellets as for the Alcator C and FTU experiments. When the profile is not peaked, the confinement time saturates as the density is increased above a value which is an increasing function of the plasma current. The values of the saturated confinement time correspond to those of the so called L-regime. Pellet injection to prevent saturation in the Alcator C experiments was suggested on the basis that Ion Temperature Gradient driven modes were responsible for most of the observed energy transport and were preventable by peaked density profile. Recent experiments by the FTU machine^2 are shown to be consistent with those performed by Alcator C originally.

^2A. Frigione et al., paper EX/P4-02, 19th Fus. Energy Conference (Lyon) 2002, publ. IAEA-CN-94.

[RP1.054] Issues for Burning Plasma Experiments and Fusion Ramp;D Metrics

Attractive magnetic fusion based on the tokamak will require significant progress in: plasma confinement for self-heating, MHD stability for high plasma pressure, self-generation of the confining magnetic field, and high power density exhaust through the plasma edge. A burning plasma experiment faces the challenge of first being able to access burning plasma conditions, and to then extend dimensional and normalized plasma parameters toward the fusion plasma regimes described in the ARIES studies. The ARIES plasma regimes also require the plasma to evolve to a largely self-organized plasma state that will present a grand challenge for magnetic fusion science. ITER, an international collaboration on a reactor-scale facility to demonstrate the scientific and technological feasibility of fusion, and FIRE, a national design study of a next step experiment to explore and optimize burning plasmas, are being pursued to address the issues described above. The dimensionless (e.g., H, beta, f_bs, Q, etc.) and normalized (e.g., fusion power density, P_loss/R, etc. ) metrics for fusion Ramp;D that can be accessed on ITER and FIRE will be described. Additional information is available at http://fire.pppl.gov. Work supported by DOE Contract # DE-AC02-76CH0 3073.

[RP1.055] Beryllium/Tungsten Mixed Material Analysis of FIRE Plasma Facing Components Including Convective Transport

Extensive computer modeling for the Fusion Ignition Research Experiment (FIRE) design study focused on Be/W mixed-material erosion issues, combining several computer codes. Since the FIRE design calls for a beryllium first wall and tungsten divertor, Be can be sputtered and transported to the divertor, forming a Be/W mixture. The goal is to determine the amount of Be deposited on divertor surfaces and to model the mixture's erosion/redeposition properties. Sputtering is calculated from deuterium neutral fluxes (obtained from DEGAS2) and D^+ flux from the plasma. The ion flux includes both the diffusive flux from the UEDGE fluid code and approximations of convective (non-diffusive) transport. The sputtering of Be is determined by VFTRIM-3D. WBC+, part of Argonne's REDEP impurity transport package, calculates the transport of Be to the divertor. Results show that convective transport dominates, increasing Be sputtering to 8x10^21 s^-1, with 4x10^21 s^-1 reaching the divertor plates. The next step is analysis of the erosion properties of the Be/W mixed material with the ITMC code at ANL.

[RP1.056] RWM Feedback Control in ITER

Achieving the advanced tokamak plasmas in ITER that are necessary to realize long-pulse/steady-state burning plasma goals depend critically on the control of the resistive wall mode (RWM) at and above the no-wall beta limit. We have carried out a broad study with the VALEN code in both eigenvalue and time-dependent analysis of RWM feeedback control using the base ITER design error field correction coils and find this system quite limited in performance, reaching only 203D analysis of the passive stabilization properties of the double wall ITER vacuum vessel and internal blanket modules, show substantial improvements in RWM ideal wall stabilization limits from the effect of the blanket modules. VALEN modeling shows this improved ideal limit can be achieved with active feedback control using only six internal coils (arranged as three n=1 pairs) with dramatic reductions in the stabilization power requirements.

[RP1.057] Calculation of beam ion distributions in ITER and their impact on alpha-particle measurements by collective Thomson scattering

Collective Thomson Scattering (CTS) has been proposed as a viable diagnostic for characterizing fusion born alpha-distributions in ITER. However, the velocities of the 1MeV proton heating beam ions in ITER are similar to that of fusion born alpha particles and may therefore mask the measurements of the fusion products. To estimate the impact that beam ions may have on the CTS measurements, a fast general code is being developed for calculation of beam ion phase-space distributions in Tokamaks. For the present ITER beam injection design and magnetic geometry the code first calculates the beam deposition profile in constant of motion (COM) space. The slowing down process is then modeled taking into account the spatially non-uniform drag by the electrons and pitch angle diffusion by the ions. Our investigations show that there are CTS scattering orientations where the alpha particles will not be significantly masked by the presence of the beam ions.

[RP1.058] Results from the US-EU Collaboration on mixed-material PMI effects for ITER

A beryllium-seeded deuterium plasma is used to conduct experiments in PISCES-B to investigate mixed-material erosion and redeposition properties of materials relevant for next step burning plasma devices. The experiments are designed to reduce uncertainties in the prediction of tritium retention in redeposited mixed-materials. A small (0.15plasma onto a carbon target is seen to have a dramatic impact on the carbon chemical and physical erosion rates. The surfaces resulting from the bombardment of graphite with D+Be plasma are measured to contain a large fraction of beryllium coverage on the surface, a result that can be explained by redeposition of eroded material. The implications of these experiments for ITER will be discussed.

Work performed as part of a USDOE-EFDA Bilateral Collaboration and supported under DOE contract DE-FG03-95ER-54301.

[RP1.059] High Power Microwaves

[RP1.060] Heating of Thin Films on High Power Microwave Windows

Thin film coatings of conductive materials with low secondary electron emission (SEE) yields, such as Ti, TiN or Cr_2O_3, are deposited on the windows to prevent the occurrence of multipactor. Contaminants may also inadvertently contribute to patches of conducting thin films on the millimeter wave diamond windows. The ohmic losses in these films can be severe, and may cause unexpected window failures. This paper provides a quantitative analysis of the power absorbed by such films, and the accompanied temperature rise. It is found that up to 50% of the incident power may be absorbed in the protective films of nanometer thickness [1]. We will also report our analysis of some recent measurements on the effects of contaminants on diamond windows [2,3]. [1] H. Bosman, Y. Y. Lau, and R. M. Gilgenbach, \textitAppl. Phys. Lett. \textbf82, 1353 (2003). [2] H. Jory, private communication (2003). [3] R. Heidinger, G. Dammertz, A. Meier, and M. K. Thumm, \textitIEEE Trans. Plasma Sci. \textbf30, 800 (2002).

[RP1.061] Development of 1 to 1.5 MW CW Gyrotrons

High power mm-wave sources provide electron cyclotron heating, current drive, and instability suppression in fusion plasmas. CPI has delivered three 110 GHz, 1 MW gyrotrons for ECH and ECCD experiments on DIII-D, each of which has demonstrated reliable operation at 1 MW for pulse lengths up to 5 seconds. CPI has also delivered a 140 GHz, 1 MW gyrotron to IPP for use on W-7X. This gyrotron has produced a peak output power of 900 kW, and pulse lengths up to 700 seconds at 500 kW. Ten consecutive 500 kW 600 second pulses were demonstrated without fault at 25CPI's test facilities could not support long pulse operation at full power, such operation is planned at IPP. The 140 GHz system employs a diode magnetron injection gun, a TE_28,7 cavity interaction mode, an internal mode converter to produce a Gaussian output beam, a low-loss CVD diamond output window, and a single-stage depressed-voltage beam collector to enhance the overall electrical efficiency of the device. Currently, under a DOE development program, CPI is fabricating a 110 GHz, 1.3-1.5 MW CW depressed-collector gyrotron to deliver improved reliability at power levels above 1 MW.

[RP1.062] Electron-cyclotron Heating on TCV with 118 GHz Gyrotrons

Three 118GHz, 0.5MW, 2 sec. gyrotrons have been installed on TCV for the purpose of electron cyclotron heating at the third harmonic in the X-mode (X3). The radiation is vertically launched. In this way, the absorption is maximized because the electromagnetic beam traverses the length of the cyclotron resonance; a curve nearly following a vertical line in the poloidal plane. The plasma facing mirror can be steered, during a pulse, in the poloidal plane and can be shifted, radially, between pulses. This allows one to control the absorption in real time during a pulse. The resonance is very narrow and coupling of the beam to the resonant electrons can be lost due to refraction and/or plasma movement. As a result it is necessary to use real time control of the launcher mirror to maintain absorption. The real time control system will be described and results from its useage will be described. The vertical launch allows the X3 system to be used as a ‘probe’ of fast electrons. By moving the radial position of the launcher from shot to identical shot it is possible to selectively heat non-thermal electrons at energies commensurate with the shift toward the high field side from the cold resonance. Experiments studying this possibility will be described. Using X3 allows access to densities above the cut-off for second harmonic ECRH. In particular X3 gives access to core heated H-mode plasmas in TCV. Results from attempts to heat H-mode using X3 will be presented and in particular, the effect of X3 heating on ELMs will be presented.

[RP1.063] An explosive emission cathode in a thermionic cathode environment

Present-day high power microwave devices suffer from a lack of reliable, reproducible cathodes for generating the requisite GW-level electron beam in a vacuum. Standard explosive emission cathodes have been limited to 10’s or 100’s of ns due to the expansion of cathode-generated plasma and the ensuing impedance collapse that debilitates microwave output. Traditional thermionic cathodes do not suffer from this drawback of plasma generation, but have not yet been demonstrated to provide the required emission current densities. It is expected that if the plasma could be made cooler and less dense, explosive emission would be more stable. Cesium iodide (CsI) has been found to slow the impedance collapse in many explosive emission cathodes. Herein we will examine diode impedance collapse, gas production, and cathode conditioning in an effort to perform an evaluation of explosive cathode performance in a typical thermionic electron gun environment. These results will then be used to help demarcate the parameter space over which these CsI-coated carbon fiber cathodes are viable candidates for the electron beam source in next-generation high power microwave devices.

[RP1.064] Secondary Electron Yield Measurements

One of the major problems affecting the efficiency of high power microwave devices is that of secodary electron emission. It is a well-known fact that efficiency can be greatly improved by using depressed collectors comprised of low yield materials. This experimental research benchmarks existing material yield curves and classifies yields of new material for use in depressed collectors. Secondary Electron Yield is defined as \delta=I_s/I_p where I_s is the secondary electron current and I_p is the primary electron current. Since I_p = I_s+I_target yield can be obtained from delta=1-I_target/I_beam. Beam current is measured using a Faraday Cup connected to pico amp meter and target current is measured directly by pico amp meter. Beam spot size is kept minimal at 1-2mm. Results show that the secondary electron yield curve is dependent upon the incidence angle of the primary electrons.

* Work supported by a grant from AFOSR

[RP1.065] Cloud Turbulence Correlation Functions and Power Spectra Measured using a Gyroklystron-Powered 94 GHz Radar

The Naval Research Laboratory (NRL) has recently developed a high power 94 GHz radar called WARLOC. This radar has unique advantages for cloud research stemming from the fact that the return from clouds scales inversely as the fourth power of the wavelength. Clouds are largely invisible to conventional radars and opaque to lidars, whereas millimeter-wave radars produce strong signals from cloud water droplets. Thus W-Band radars can be used to sense the internal structure of clouds. The WARLOC transmitter has about three orders-of-magnitude more average power than the W-Band radars used in previous cloud studies and greatly improved resolution and scanning capability. Here we report initial results on cloud studies. The new capabilities of WARLOC have allowed us to produce high-resolution images of the internal structure of clouds. Regions many square kilometers in area can be scanned with 15 m resolution in about a minute even through intervening cloud layers. The scanned cloud reflectivity yields two-dimensional cloud turbulence correlation functions and power spectra directly from spatial measurements for the first time, and with higher resolution than previously possible. We find that in the inertial range, the Kolmogorov spectral index (-5/3) agrees reasonably well with the data, but the assumption of isotropy does not. Interestingly, in two clouds studied, at longer scale lengths, the fluctuations appear to be wavelike in the vertical direction, but not in the horizontal direction.

[RP1.066] Microwave Noise Reduction Experiments in kW Magnetrons

Nonrelativistic magnetron (4 kV, 300 mA, 2.45 GHz, kW microwave power) experiments are performed on commercial oven magnetrons for an in-depth investigation of crossed-field injection-locking and noise. Injection-locking is demonstrated by utilizing an oven magnetron as a reflection amplifier. Microwave noise generation is explored as a function of injected signal and cathode conditions. Our experiments confirm that the microwave noise can be reduced when the cathode heater power is decreased. We also investigate the influence of reducing the cathode heater power on the microwave output power and magnetron efficiency. Noise reduction is also observed when a weak, low noise microwave signal is injected into the reflection amplifier. Experimental and simulation results will be reported for innovative, new techniques for reduction of microwave noise in magnetrons.

[RP1.067] Analysis of Mode Growth in Relativistic Magnetrons with Extractor Loading

Relativistic magnetrons are a promising source of high power microwaves, combining high power and repetition rate with the potential for high efficiency. Although they have been actively investigated for over three decades, much of the theoretical work has focused on the performance of the source itself, typically without the complication of power extraction schemes. Experimentally, however, it has been observed that the loading of cavities due to extraction can greatly increase the potential for mode competition, as well as complicate the resonant structure of the device. In this work, we develop a circuit model for a relativistic, cylindrical, N-cavity magnetron with M extractors. In order to predict the growth rates of the modes the beam is prescribed as either a cycloidal electron flow or Brillouin electron flow. The impact of these two models for the electron dynamics with be compared and contrasted. Detailed analysis of the hot tube, small signal, growth rates for the magnetron modes will be examined with both a simple analytic space charge model and a semi-empirical model found from 3-D particle-in-cell simulation. Implications on extractor design, mode competition, and optimization of the relativistic magnetron will be discussed.

[RP1.068] Initiation of a Relativistic Magnetron

We report on recent results in our studies of relativistic magnetrons. Experimentally, these devices have proven to be very difficult to operate, typically cutting off too quickly after they are initialized, and therefore not delivering the power levels expected [1]. Our analysis is based on our model of a crossed-field device, consisting only of its two dominant modes, a DC background and an RF oscillating mode [2]. This approach has produced generally quantitatively correct values for the operating regime and major features of nonrelativistic devices. We have performed a fully electromagnetic, relativistic analysis of a magnetron of the A6 cylindrical configuration. We will show that when the device should generate maximum power, it enters a regime where the DC background could become potentially unstable. In particular, when a nonrelativistic planar device enters the saturation regime, the DC electron density distribution could become unstable if the vertical DC velocity would ever become equal to the magnitude of the vertical RF velocity [3]. We find that during the initiation phase, for the highest power levels of our model of the A6, near the cathode, the DC vertical velocity does become just less than, and definitely on the order of the magnitude of the vertical RF velocity. Consequently, any localized surge in the currents near the cathode, could easily destroy the smooth upward flow of the electrons, drive the DC background unstable, and thereby shut down the operation of the device.

[1] Long-pulse relativistic magnetron experiments, M.R. Lopez, R.M. Gilgenbach, Y.Y. Lau, D.W. Jordan, M.D. Johnston, M.C. Jones, V.B. Neculaes, T.A. Spencer, J.W. Luginsland, M.D. Haworth, R.W.Lemke, D. Price, and L. Ludeking, Proc. of SPIE Aerosense 4720, 10-17, (2002). [2] Theoretical modeling of crossed-field electron vacuum devices, D.J. Kaup, Phys. of Plasmas 8, 2473-80 (2001). [3] Initiation and Stationary Operating States in a Crossed-Field Vacuum Electron Device, D. J. Kaup, Proc. of SPIE Aerosense 4720, 67-74, (2002).

[RP1.069] Mode Control, RF Breakdown and Innovative Cathodes in Long-Pulse, Relativistic Magnetron Experiments

Research is underway on a 6-vane, Titan relativistic magnetron powered by MELBA-C (Ceramic insulator) at parameters: V = -0.3 MV, I = 1-10 kA, and \tau = 0.5 \mus at 10^-7 Torr. This experiment has generated peak microwave power levels of 300-500 MW near 1 GHz. Time-Frequency Analysis (TFA) indicates mode competition between pi mode and 2/3 pi mode during operation. Experiments to investigate the effect of extractor symmetry on mode competition have utilized two vs three waveguides distributed symmetrically on the magnetron structure. In agreement with AFRL simulations three-fold symmetry tends to improve pi mode operation. Ongoing research involves efforts to prime this device with a MW magnetron source. RF breakdown has been observed in a window/waveguide system filled with atmospheric pressure air. Microwave power and pulselength are both increased with SF6 fill gas. Innovative, new, high-current cathodes are being explored utilizing laser-ablation.

* Research supported by AFOSR and AFRL.

[RP1.070] Vortex Structure in a High-Density Non-Axisymmetric Crossed-Field Equilibrium Flow

It is shown that vortex structures exist in high-density non-axisymmetric crossed-field equilibrium flow in a crossed-field device. With the aid of perturbation theory and the guiding-center approximation and assuming a constant electron particle number density in the electron layer, we determine equipotential surfaces, electron trajectories, and the corrugated boundary of the electron layer. As the electron density approaches the (critical) Brillouin density, a vortex structure with regions of negative potential forms near the cathode surface.

[RP1.071] Compact Relativistic Magnetron with Output Mode Converter

We consider a relativistic magnetron in which all of the resonators of the anode block are smoothly continued onto a conical antenna up to the radius corresponding to the cutoff frequency of the radiated wave in a cylindrical waveguide. Such a magnetron is capable of high output power, is compact, has a high resistance to microwave breakdown, is able to work with extremely high currents, and has the possibility of forming desirable output radiation patterns. The magnetic field can be provided by a small solenoid over the resonant system, which is a much smaller volume than is required for the Helmholtz coils used in traditional relativistic magnetrons. The maximum size of this magnetron is the aperture of the horn antenna. The unique aspect of such a design is the possibility of using the horn antenna for conversion of the operating mode to lower order modes, including the TE_11 mode, which is radiated as a narrow wave beam. For a magnetron operating in \pi-mode, the mode converter comprises a continuation of the resonantor blocks onto the horn for those resonators that correspond to the symmetry of the output mode. For example, in order to provide Gaussian mode output only two diametrically opposite resonators of even-numbered resonators must be continued onto the horn. In this case the aperture of the horn antenna can be close to the cut-off diameter for the TE_11 mode, and the output power is limited only by breakdown of the output window. In this presentation results of preliminary calculations of the magnetron with output mode converters are presented.

[RP1.072] Laminar Electron Flow Solutions for Two-Dimensional Electrode Geometries

Numerous microwave devices (such as magnetrons and sheet beam devices) utilize geometries which are invariant in one dimension. In order to minimize noise and beam loss, one seeks conditions under which two-dimensional laminar electron flows may exist in such devices. In this work, a non-relativistic fluid equilibrium theory for these electron flows is developed which takes into account self-fields, image fields from the electrodes, and externally imposed electric and magnetic fields. Several special cases are examined within the theory, including the Brillouin flow limit.

[RP1.073] Three Dimensional Modeling Of Electron Guns Using MICHELLE

MICHELLE, which is a new two-dimensional (2D) and three-dimensional (3D) electrostatic equilibrium particle-in-cell (PIC) code, has been designed to address the recent beam optics modeling and simulation requirements for vacuum electron devices, ion sources, and charged-particle transport. Problem classes specifically targeted include gridded-guns, multi-beam guns, sheet-beam guns, and depressed collectors. The focus of the development program is to combine modern finite-element techniques with improved physics models. The code employs a conformal mesh, including both structured and unstructured mesh architectures for meshing flexibility, along with a new method for accurate, efficient particle tracking. New particle emission models for thermionic beam representation are included that support primary emission, and secondary emission is handled with an advanced model. One of the key features of the MICHELLE code is its ability to model fine-scale features in a large volume. This has made it valuable for performing sensitivity studies and generating manufacturing tolerance specifications. The MICHELLE code has been released, and has been employed in the design and verification of electron guns and collector designs over the last several years. Much emphasis has been placed on validating the code against other models and experimental data. An open issue for validation concerns secondary emission. MICHELLE has a new extensive secondary emission model, however data is sparse for many beam energies and materials of interest. The application of MICHELLE to multi-beam guns, gridded guns, and multistage depressed collectors will be presented, along with an update of the MICHELLE development regarding the areas of particle tracking, temperature-limited and space-charge-limited emission, and secondary emission.

[RP1.074] Numerical modeling of breakdown in high-power waveguides

Unwanted electron emission from metal walls influences the operation of vacuum waveguides. If the electron density becomes large enough, the electron plasma can reflect the electromagnetic waves that are supposed to be transmitted. Numerical simulations of the generation of electrons in high-power vacuum waveguides are presented. In particular, the parameters relevant to the high-power waveguides at the Stanford Linear Accelerator Center (SLAC) are modeled. Modeling is done with the VORPAL code from the University of Colorado. Results are shown comparing experiments from SLAC to a thermal gas desorption and ionization model. Researchers have used this model to successfully explain results from ion diode breakdown.

[RP1.075] Ion Beams

[RP1.076] Recent results from the High Current Experiment for Heavy Ion Inertial Fusion

The High Current Experiment (HCX) is exploring heavy-ion beam transport in an alternating gradient quadrupole focusing channel at a scale representative of the low-energy end of an induction linac driver for fusion energy production. A primary mission of this experiment is to investigate aperture fill factors (F = diameter of maximum beam excursion / aperture diameter) acceptable for the transport of space-charge dominated heavy-ion beams at high space-charge intensity (line-charge density \sim 0.2\mu C) over long pulse durations (> 4 microsec). We present the phase space evolution of a K+ ion beam transported (F>0.5) through the matching section, 10 electrostatic transport quadrupoles and 4 magnetic quadrupoles. Transverse phase space is measured including beam halo and particle loss. Space charge waves are also studied along with their influence on the transverse beam evolution. New data on beam energy and phase space projections better constrain beam parameters and improve agreement with simulations. The consequences for beam control and fill-factors in future induction accelerators will be discussed.

[RP1.077] Progress in Measuring Electron Cloud Effects in HIF Accelerators

Accelerators for heavy-ion inertial fusion energy (HIF) have an economic incentive to fit beam tubes tightly to beams, putting them at risk from electron clouds produced by secondary electrons, and ionization of gas from walls (a problem in accelerators for high-energy physics). We are using two complementary approaches with the High-Current Experiment (HCX): (1) Measure the electron emission and gas desorption coefficients from 1 MeV K^+ ions incident on a target, at angles near grazing incidence, to understand the processes and develop mitigation techniques. (2) Characterize electron production, accumulation, and the effects on ion beams in quadrupole magnets where we are commissioning a variety of charged particle diagnostics to measure net charge, gas density, and secondary electron production. These data will be compared with predictions of theory and simulations.

[RP1.078] Recent developments in diagnostics for Heavy Ion Fusion experiments

We discuss progress in diagnostic development for the Heavy Ion Fusion program in the HIF-VNL at LBNL and LLNL. Typical HIF beams are high current (up to 1 A), and the energy range is 60 keV to 2.0 MeV, increasing up to 10 MeV in the near future. Beam parameters of interest include current, density distribution, energy, energy distribution, emittance, and space potential, in injector, transport, and final focus sections. Optical diagnostics based on a scintillator screen and a gated intensified CCD camera have been implemented and provide full 4-D transverse information on the experimental beams. Current work includes development of a compact optical diagnostic and improved algorithms for data analysis and interpretation. A longitudinal diagnostic kicker has been implemented for generating longitudinal space-charge waves. Comparison of the waves with a simple 1-D fluid model of the beam will be presented. Time of flight of the space charge wave and an electrostatic energy analyzer provide an absolute measure of the beam energy. These and other new diagnostics will be described.

[RP1.079] Progress in Modeling Electron Cloud Effects in HIF Accelerators

Stray electrons can arise in positive-charge accelerators for heavy ion fusion (or other applications) from ionization of gas (ambient or released from walls), or via secondary emission. Their accumulation is affected by the beam potential and duration, and the accelerating and confining fields. We present electron orbit simulations which show the resultant e-cloud distribution; ion simulations with prescribed e-clouds which show the effect on ion beam quality; a gyro-averaged model for including electron dynamics in ion simulations, and its implementation status; and progress in merging the capabilities of WARP (3-D PIC code for HIF) (D.P. Grote, A. Friedman, I. Haber, Proc. 1996 Comp. Accel. Physics Conf., AIP Proc. 391), 51 (1996), with those of POSINST (e-clouds in high-energy accelerators) (M.A. Furman, LBNL-41482/CBP Note 247/LHC Project Report 180, May 20, 1998).

[RP1.080] The CMEE Library for Numerical Modeling of Electron Effects

The CMEE (Computational Modules for Electron Effects) library is a collection of computer routines for numerical modeling of electron effects in accelerator and plasma physics codes. The goal of this library is to make these numerical models available to any code in need of electron effects modeling, including high-power microwave codes, fusion wall interaction codes, laser-plasma codes, proton accelerator codes, and HIF codes. CMEE includes routines to model secondary electrons, neutral gas desorption and ionization. The secondary electron routines are based on routines from the POSINST code. The neutral gas desorption routines are based on a thermal binding model similar to the model in ,e.g., the LSP code. The ionization routines are based on the IONPACK library from Tech-X. This poster discusses the latest state of these routines, specifically implementation in the WARP code and comparisons to data from the High Current Experiment (HCX). In particular, recent comparisons between the CMEE routines and neutral gas desorption measurements from HCX are presented.

[RP1.081] The Integrated Beam Experiment-- The Next Step for Heavy Ion Fusion

The U.S. Heavy Ion Fusion Program is completing a 2-decade effort of small-scale experiments and analysis to explore the beam manipulations and non-neutral plasma physics of the intense beams necessary for a driver. The next step, a proof-of-principle experiment called the “Integrated Beam Experiment” (IBX) is in the design stage. It would integrate, for a single beam, all the beam physics from source to target, including beam production, acceleration, interaction with electrons and gas, compression, neutralization, and final focus. Present designs call for a K+ beam accelerated in an induction linac to 5-10 MeV. We present design studies, including PIC code studies of beam behavior.

[RP1.082] Simulation Studies of Temperature Anisotropy Instability in Intense Charged Particle Beams for IBX Parameters

The Integrated Beam Experiment (IBX) is a proof-of-principal experiment for heavy ion fusion designed to test source-to-target beam physics using a single beam of K^+ ions of duration 0.2 - 1.5 \mu s, accelerated to energies ~ 5-10 MeV, and driver-scale normalized perviance in the range 10^-5 -10^-3. An important physics issue to be addressed by IBX is the effect of longitudinal-transverse coupling on the beam transport and focusibility of the driver. Our previous numerical and theoretical studies of intense charged particle beams with large temperature anisotropy [E. A. Startsev, R. C. Davidson and H. Qin, Phys. Plasmas \textbf9, 3138, 2002] demonstrated that a fast, electrostatic, Harris-like instability may develop. This paper reports the results of recent simulations of the temperature anisotropy instability using the Beam Equilibrium Stability Transport (BEST) code for IBX parameters.

[RP1.083] Moment Equation Approach to Chromatic Aberrations in Final Focus Systems for Heavy Ion Fusion

Ordinarily, the envelope equations provide adequate accuracy for rapid design calculations of rough layouts of a set of magnetic quadrupoles arranged to focus an ion beam to a small spot (for example, for the final focus system of a Heavy Ion Fusion driver.) Typically, such systems consist of four to eight large aperture quadrupole magnets preceded by a drift section. However, chromatic aberrations (depending on momentum spread) and other non-linear terms in the equations of motion may contribute significantly to the final focal spot size. We present here a moment equation approach which includes terms through second order in particle transverse position, angle and fractional momentum spread, (but which drops terms higher than a specified order) and which treats space charge by assuming it is distributed uniformly over an elliptical cross section. Integrating the derived set of moment equations allows rapid determination of the contribution of chromatic aberrations to spot size. We compare the results of integrating moment equations with an analytic theory, as well as with Particle-In-Cell simulations.

[RP1.084] RF Plasma Source for Heavy Ion Fusion

We are developing high-current ion sources for Heavy Ion Fusion (HIF) applications. Heavy ion driven inertial fusion requires beams of high brightness to deposit the necessary high energy in the target to obtain high gain. Our proposed RF plasma source starts with an array of high current density mini-beamlets (of a few mA each at \sim100 mA/cm2) that are kept separated from each other within a set of acceleration grids in order to minimize the space charge expansion. After they have gained sufficient kinetic energy (>1.2 MeV), the mini-beamlets will be allowed to merge together to form a high current beam (about 0.5 A) with low emittance. We are performing experiments on RF plasma sources. A 80-kV 20-µs source has produced up to 5 mA of Ar+ in a single beamlet. We have measured the emittance of a beamlet, and the fraction of Ar++ ions. The plasma chamber has 26-cm inner diameter with multicusp permanent magnets to confine plasma. RF power (\sim11 MHz, >10 kW) is applied to the source via a 2-turn, 11-cm diameter antenna inside the chamber. We have started testing a 80-kV 61-hole multi-beamlet array designed to produce a total current >200 mA. In this stage of the experiments the beamlets will not be merged into a single beam. A 500-kV experiment where the beamlets will be merged to a produce 0.5 A beam is being planned.

[RP1.085] Particle Simulation Schemes for High Intensity Charged Particle Beams

Numerical schemes for the electromagnetic particle simulations of high intensity charged particle beams have been developed. The purpose of devising these schemes is to avoid the numerical difficulties associated with the direct calculation of the time derivatives of the vector potential, \partial A / \partial t, in the Darwin model, for which the transverse induction current in Ampere's law is neglected. The first scheme requires the calculations of higher order velocity moments of the distribution function to obtain the time derivatives for both the scalar potential \Phi and A, similar to the method used for shear-Alfven waves.[1] The second uses the canonical momentum P = p + q A/c in the equations of motion as a means to eliminate the troublesome time derivatives.[2] The use of these schemes for physics problems in heavy ion fusion systems will be reported.

[1] W. W. Lee, J. L. V. Lewandowski, T. S. Hahm, and Z. Lin, Phys. Plasmas 8, 4435 (2001).

[2] W. W. Lee, E. Startsev, H. Qin and R. C. Davidson, Proceedings of 2001 Particle Accelerator Conference 1906 (2001).

[RP1.086] Beam Loss and Halo Formation Induced by Image-Charge Effects in a Small-Aperture Alternating-Gradient Focusing System

It is shown with a test-particle model that image-charge effects induce a new mechanism for chaotic particle motion and halo formation in an intense charged-particle beam propagating through an alternating-gradient focusing channel with a small aperture, circular, perfectly conducting pipe. While our model allows for nonuniform beams with elliptic symmetry, the effects of image charges on halo formation are illustrated with a uniform Kapchinskij-Vladimirskij (KV) distribution. Halo formation and chaotic particle motion are studied for various choices of system parameters: filling factor, perveance and vacuum phase advance. Furthermore, the percentage of beam loss to the conductor wall is calculated as a function of propagating distance and aperture size.

[RP1.087] Ionization Cross Sections for Ion-Atom Collisions in High Energy Ion Beams

Knowledge of ion-atom ionization cross sections is of great importance for many accelerator applications. When experimental data and theoretical calculations are not available, approximate formulas are frequently used. Based on experimental data and theoretical predictions, a new fit for ionization cross sections by fully stripped ions is proposed. The Born approximation and classical trajectory calculations are frequently used to estimate the cross sections. Neither approximation is expected to be valid over the entire range of projectile ions and target atoms. Aspects of both models must be included in order to address the shortcomings in the underlying assumptions. A large difference in cross section, up to a factor of six, calculated in quantum mechanics and classical mechanics, has been obtained for 3.2GeV negative ions of iodine and positive ions of cesium. Because at such high velocities the Born approximation is well validated, the classical trajectory approach fails to correctly predict the stripping cross section at high energies for electron orbitals with low ionization potential.

[RP1.088] Symmetric Neutralized Ion Beams - Experiments and PIC Simulation

Experiments with symmetric neutralized ion beams consisting of H^+ and H^- ions are being conducted. The H^+ and H^- beams are transported and merged in RF quadrupoles and are then injected into a magnetic field. The ability of the beam to propagate across the magnetic field due to the formation of a polarization electric field that cancels the magnetic field deflection is being explored. Approximate quantities for the proof-of-principle experiment are: beam energy 1 keV, beam current density 10^9 mA/cm^2, magnetic field 100 G. 3D PIC simulation of the cross-field propagation process is also being carried out, with particular attention to the effect of a background plasma and other loss mechanisms such as beam expansion. The injection of arrays of higher energy beams of this type into magnetic fusion devices for heating, fueling, current drive, and diagnostics is being considered.

[RP1.089] RF Plasma Source for Heavy Ion Beam Charge Neutralization

Highly ionized plasmas are being employed as a medium for charge neutralizing heavy ion beams in order to focus to a small spot size. Calculations suggest that plasma at a density of 1 - 100 times the ion beam density and at a length ~ 0.1-0.5 m would be suitable for achieving a high level of charge neutralization. An ECR source has been built at the Princeton Plasma Physics Laboratory (PPPL) in support of the joint Neutralized Transport Experiment (NTX) at the Lawrence Berkeley National Laboratory (LBNL) to study ion beam neutralization with plasma. The ECR source operates at 13.6 MHz and with solenoid magnetic fields of 0-10 gauss. The goal is to operate the source at pressures ~ 10^-5 Torr at full ionization. The initial operation of the source has been at pressures of 10^-4 - 10^-1 Torr. Electron densities in the range of 10^8 - 10^11 cm^-3 have been achieved. Recently, pulsed operation of the source has enabled operation at pressures in the 10^-6 Torr range with densities of 10^11 cm^-3. Near 100% ionization has been achieved. The source has been integrated with NTX and is being used in the experiments. The plasma is approximately 10 cm in length in the direction of the beam propagation. Modifications to the source will be presented that increase its length in the direction of beam propagation.

[RP1.090] Modeling Drift Compression in an Integrated Beam Experiment for Heavy-Ion-Fusion

The Integrated Beam Experiment (IBX) is an induction accelerator being designed to further develop the science base for heavy-ion fusion. The experiment is being developed jointly by Lawrence Berkeley National Laboratory, Lawrence Livermore National Laboratory, and Princeton Plasma Physics Laboratory. One conceptual approach would first accelerate a 0.5-1 A beam of singly charged potassium ions to 5 MeV, impose a head-to-tail velocity tilt to compress the beam longitudinally, and finally focus the beam radiallly using a series of quadrupole lenses. The lengthwise compression is a critical step because the radial size must be controlled as the current increases, and the beam emittance must be kept minimal. The work reported here first uses the moment-based model HERMES to design the drift-compression beam line and to assess the sensitivity of the final beam profile to beam and lattice errors. The particle-in-cell code WARP is then used to validate the physics design, study the phase-space evolution, and quantify the emittance growth.

[RP1.091] A Proposal for an Initial Fusion Materials Testing Facility

As pointed out in the Fusion Community "35 Year Plan" presented at the APS DPP in 2002 in Orlando, fusion materials that are low activation, have long lifetimes and can withstand high neutron fluxes are essential for fusion energy for both MFE and IFE fusion power systems. The present US 35 year proposed strategy for fusion materials development calls for building a test facility within the next ten years. Initial estimates put the cost of the International Fusion Material Irradiation Facility (IFMIF) at nearly $1B. We report the possibility of an initial, smaller scale fusion material testing facility built for a significantly lower cost utilizing the existing infrastructure of the $150M LEDA facility located at LANL. While the test volumes would be much smaller than the full scale IFMIF, there are significant advantages to beginning a smaller scale testing program sooner than the planned start of operations for the full scale IFMIF.

[RP1.092] Preliminary Results of Proton Collimation Experiments

Proton Collimation involves the conversion of an isotropic source of highly energetic protons from a source, such as an inertial electrostatic confinement fusion reactor, and forcing the particles into a highly focused channel where they can be utilized in a direct energy converter as a possible spacecraft power source or possibly used as a high-Isp, low-thrust particle rocket. These experiments are to verify the proton collimator effect, its reciprocal process, and to provide an evaluated database for use in the design of a proton collimator for an IEC fusion reactor. A low energy electron beam is used to simulate an energetic fusion proton beam in the current experiments. Sensitivity experiments will provide the database needed for constructing an optimized collimator. In particular, the effect of collimator operation to source asymmetry will be tested using the adjustable-position source feedthrough. In the process of completing the collimator construction and testing, several minor modifications, such as filament structural design, have been made. Testing of device settings and collimator principles is underway and initial results will be presented.

[RP1.093] ICF Hohlraums and Implosion Physics

[RP1.094] Investigation of M-band Gold Spectra in Hot Halfraums at HELEN and OMEGA and in EBIT plasmas

The determination of the charge state distribution (CSD) of highly ionized gold in high-density plasmas (\sim 10^21 cm^-3) is critical for ICF experiments. Predictive calculations of the CSD have produced widely varying results. Previous experiments on NOVA have had some success in guiding the models at 2 keV. Here, CSD’s have been investigated in Hot Halfraum experiments (T_e > 5keV) at OMEGA and HELEN by comparing the n=5-3 and n=4-3 X-ray transitions in Ni-like to Kr-like Au to HULLAC line emission modeling. CSDs at low density (\sim 10^12 cm^-3) and T_e = 2-2.5 keV have been inferred in plasmas created at the Livermore electron beam ion trap (EBIT) facility (Wong, et. al., PRL 90, 235001-1 (2003)). One uncertainty in these CSD measurements is the collisional excitation cross sections. To reduce this uncertainty, some of the relevant n=3-4 and n=3-5 cross sections have been measured at the EBIT facility and are compared with theoretical predictions. This work was performed by the University of California LLNL under the auspices of the DOE under contract W-7405-ENG-48..

[RP1.095] Simulations of the effects of foam ablators on shock propagation and implosion stability in ICF capsules.

We have performed a series of detailed simulations of shock propagation through two-dimensional models of foam ablators, using NRL's FAST radiation hydrodynamics code (J.H. Gardner, A.J. Schmitt, et al., Phys. Plasmas 5, 1935 (1998)). These models differ from those used in previous simulations in incorporating the effects of radiation, heat, and detailed equations of state for the components of the foams (CH, DT, and RF). It was found that the selective absorption of radiation by the CH or RF fibers dramatically changes the structure of the foam in front of the shock, which in turn has a large effect on the characteristics of the propagating shockwave. We attempt to draw some preliminary conclusions for the design of high-gain direct-drive ICF targets. We also reexamine the results of experiments, performed at NRL's Nike laser facility, on the propagation of shocks through foam ablators, in light of our simulation results.

[RP1.096] Experiments on Ablation Front Features

Structures at a non-accelerated ablation front evolve in a Richtmyer-Meshkov-like manner, but differently from those shocked at an embedded interface [Velikovich et al., Phys. Plasmas 5, 1662 (2000); Robey et al., Phys. Rev. Lett. 8908, 5001 (2002)]. We report on simulations and experiments on nominally hemispherical features initially imposed on a TiO_2 foam, density 1.8 g/cc. The features were ablatively driven with x-rays from a gold hohlraum, 2.26 mm diameter by 1.68 mm long. The radiation temperature peaked at about 140 eV, for about 3.3 ns. The features were typically 50 \mum diameter by 10-20 \mum deep. The initial studies described concentrated on diagnostic techinques, including addressing the issues involved in backlighting along the axis of a hohlraum during the stagnation of gold on axis.

[RP1.097] Interpretation of X-ray conversion efficiency in hohlraums on Omega

Since 1999, CEA performs experiments on Omega in collaboration with LLNL. One of these experiments (may '03 empty hohlraum shots with 3 cone LMJ-like irradiation) is interpreted here. In order to understand former symmetry experiments, the target is oriented so that the backscattering diagnostics are along the inner cone. The influence of the polyimide foil holding the foam ball in the center of the hohlraum is observed.

[RP1.098] Diagnosing Dynamic Hohlraums Using Tracer X-ray Lines

A dynamic hohlraum is a unique, nearly Planckian x-ray source that is produced by imploding an array or arrays of high atomic number wires onto a cylindrical hydrocarbon foam. Its ability to generate a desired x-ray spectrum symmetrically from both the top and bottom of the foam depends critically on the plasma conditions attained in its interior. In a series of experiments performed on the Z generator, Al and/or Mg tracer layers were embedded in the foam target at various distances from its ends. Their K-shell lines were observed mostly in absorption. In the present work, we show how the relative intensities and depths of these absorption lines are controlled by the foam plasma temperature, density, their gradients, and top/bottom symmetry. Calculations using detailed configuration atomic physics and radiation transport models with up to 4000 frequency groups are employed to diagnose specific Z experiments by analyzing their tracer spectra.

[RP1.099] Improved drive symmetry via integrated 3D simulations of NIF ignition targets

The baseline NIF ignition target design has been improved with the aid of 3-D integrated HYDRA simulations, incorporating a number of modifications to accommodate realistic illumination geometry. We describe these refinements, which include a laser pointing geometry optimized to reduce the m=4 azimuthal variation associated with the inner laser ring. Integrated simulations show the capsule implosion exhibits improved symmetry and greater overall robustness than previous designs. We will also simulate the coupled effects of multimode surface perturbations and intrinsic drive asymmetry in high-resolution capsule only simulations. Perturbations initialized on both the inner and outer surfaces encompass the full spectrum of the most dangerous modes, yielding the most complete model of the ignition capsule performance.

[RP1.100] Investigation of radiation diffusion in a thin Au wall using x-ray backlit imaging

Radiation diffusion occurs in the hohlraum wall in Inertial Confinement Fusion (ICF). ICF hohlraums come in both thick and thin wall versions. The thick wall hohlraum contains the laser produced drive radiation efficiently, but precludes direct observation of the laser-wall interaction region The thin wall hohlraum (TWH) allows an x-ray image of the interaction region to be observed through the wall, with minimal impact on the hohlraum energetics. The TWH typically consists of a 2-3 um Au layer surrounded by a supporting wall of plastic or Al. Energy loss by diffusion through the thin Au layer can have a significant impact on the energetics of the hohlraum, but the losses have never been directly examined. The present work examines the radiation diffusion through a 1 um Au wall with a 25 um Al support layer by imaging the expansion of the Al under the influence of the radiation diffusing through the Au layer. 6.7 keV x-ray backlit imaging is used to examine the outer Al surface location as a function of time, and to compare the resultant trajectory to that predicted by a radiation hydrodynamics code. The experiments are done on the Sandia National Laboratories Z accelerator, using the dynamic hohlraum as a radiation source to drive the expansion of a small TWH. Images and simulations will be compared.

[RP1.101] Electron Transport In Magnetized Laser Hohlraum Plasmas

The plasma conditions we expect to encounter in hohlraum experiments on the National Ignition Facility span from the familiar, relatively benign conditions of Nova and Omega to much more hostile environments in which electron temperatures may be 10s of keV. In any of these situations simple estimates indicate that heat flow is very non-local, making the local diffusion approximation typically used in hydrocodes invalid. However, large, mega gauss magnetic fields which are generated