Search

Induction accelerators are appealing for heavy-ion driven inertial fusion energy (HIF) because of their high efficiency and their demonstrated capability to accelerate high beam current (≥10 kA in some applications). For the HIF application, accomplishments and challenges are summarized. HIF research and development has demonstrated the production of single ion beams with the required emittance, current, and energy suitable for injection into an induction linear accelerator. Driver scale beams have been transported in quadrupole channels of the order of 10% of the number of quadrupoles of a driver. We review the design and operation of induction accelerators and the relevant aspects of their use as drivers for HIF. We describe intermediate research steps that would provide the basis for a heavy-ion research facility capable of heating matter to fusion relevant temperatures and densities, and also to test and demonstrate an accelerator architecture that scales well to a fusion power plant.

We have measured electron and gas emission from 1 MeV K + impact on surfaces near grazing incidence on the High-Current Experiment (HCX) at LBNL. Electron emission coefficients reach values of 130, whereas gas desorption coefficients are near 10 4 . Mitigation techniques are being studied: a bead-blasted rough surface reduces electron emission by a factor of 10 and gas desorption by a factor of 2. Diagnostics are installed on HCX, between and within quadrupole magnets, to measure the beam halo loss, net charge and expelled ions, from which we infer gas density, electron trapping, and the effects of mitigation techniques. Here we discuss a new diagnostic technique that measures gas pressure and electron ionization rates within quadrupole magnets during the beam transit.

W.I-07 EXPERIMENTAL STUDY OF THE TRANSPORT LIMITS OF INTENSE HEAVY ION BEAMS IN THE HCX 1 L. R. Prost (FNAL), F. M. Bieniosek (LBNL), C.M. Celata (LBNL), A. Faltens (LBNL), P. A. Seidl (LBNL), W. L. Waldron (LBNL) R. Cohen (LLNL), A. Friedman (LLNL), M. Kireeff Covo (LLNL), S.M. Lund (LLNL), A.W. Molvik (LLNL) I. Haber (UM) FNAL: Fermi National Accelerator Laboratory, Batavia, IL, 60510 USA LBNL: Lawrence Berkeley National Laboratory, Berkeley, CA, 94720-8201 LLNL: Lawrence Livermore National Laboratory, Livermore, CA 94550 USA UM: University of Maryland, College Park, MD 20742-3511 Corresponding author: Lionel Prost FERMILAB P.O. Box 500 – MS 306 Batavia, IL, 60510 USA Tel: Fax: E-mail: lprost@fnal.gov Abstract The High Current Experiment (HCX) at Lawrence Berkeley National Laboratory is part of the US program to explore heavy-ion beam transport at a scale representative of the low-energy end of an induction linac driver for fusion energy production. The primary mission of this experiment is to investigate aperture fill factors acceptable for the transport of space-charge-dominated heavy-ion beams at high space-charge intensity (line charge density up to ~ 0.2 µC/m) over long pulse durations (4 µs) in alternating gradient focusing lattices of electrostatic or magnetic quadrupoles. The experiment also contributes to the practical baseline knowledge of intense beam manipulations necessary for the design, construction and operation of a heavy ion driver for inertial fusion. This experiment is testing transport issues resulting from nonlinear space-charge effects and collective modes, beam centroid alignment and beam steering, matching, image charges, halo, electron cloud effects, and longitudinal bunch control. We first present the results for a coasting 1 MeV K + ion beam transported through the first ten electrostatic transport quadrupoles, measured with optical beam-imaging and double-slit phase-space diagnostics. This includes studies at two different radial fill factors (60% and 80%), for which the beam transverse distribution was characterized in detail. Additionally, beam This work performed under the auspices of the U.S Department of Energy by University of California, Lawrence Livermore and Lawrence Berkeley National Laboratories under contracts No. W-7405-Eng-48 and DE-AC03-76SF00098.

The High-Current Experiment (HCX) at Lawrence Berkeley National Laboratory is part of the US program to explore heavy-ion beam transport at a scale representative of the low-energy end of an induction linac driver for fusion energy production. Four pulsed magnetic quadrupoles are being used to study gas and electron effects with a 0.2 A, 1-MeV K + beam. The magnets, originally designed and built for a prototype pulsed magnetic quadrupole array, have an elliptical beam tube (6×10 cm) and iron yoke. The magnet coil and field length are ≈31 cm, and operating gradients are 10–40 T/m. To establish that the field quality of the prototype quadrupoles is satisfactory for the experiments, a 1-cm pickup loop was used to measure the flux B r ( θ ) at the magnet mid-plane and also at the lead and return ends. A longer probe was used to measure the integrated flux of B θ ( θ ) along the magnet. The field quality appears satisfactory for the short transport experiments through these quadrupoles.

When a short-duration, small-amplitude energy perturbation is applied to a heavy-ion beam, positive- and negative-going longitudinal space–charge waves are generated on the beam. Longitudinal diagnostic kickers that provide ∼1% energy perturbation have been implemented for generating space–charge waves on HCX and NTX beams. The kickers consist of specially designed fast pulse generators combined with an existing aperture or ESQ structure to provide the longitudinal perturbation to the beam. The amplitudes of the resulting density waves are ∼10%, measured a few meters downstream of the kicker. The time-of-flight (TOF) of the wave provides an accurate measure of the beam energy. The TOF measurements will be described. Comparison of measured waves with a simple 1-D fluid model of the beam will be presented.

In this paper we report on two Heavy Ion Fusion (HIF) driver point design studies. The Robust Point Design (RPD) was completed over a year ago, and the Modular Point Design (MPD) is still in progress. The goal of any point design study is to construct a detailed design that is self-consistent and integrated from injector to target. This has been the primary theme of both studies.

In the planned heavy-ion Integrated Research Experiments (IRE), final beam kinetic energy will be low compared to a driver, giving perveance significantly larger than the driver, and making final focus to a small spot comparatively much more difficult. Splitting of beams ahead of the final focus system would allow economical transport of large beams through the IRE accelerator, while maintaining focusability. Since splitting decouples the number of beams transported in the accelerator from what is required at final focus, it could also lower the cost of a driver. In this paper we consider the effects on beam quality of transverse beam splitting. The system of most interest consists of one or more consecutive magnetic septa. PIC simulations of this case are discussed, along with engineering considerations.

An overview of experiments is presented, in which the physical dimensions, transverse emittance and generalized perveance are scaled to explore driver-relevant beam dynamics. Among these are beam merging, focusing to a small spot, and bending and recirculating beams. The Virtual National Laboratory for Heavy Ion Fusion (VNL) is also developing two driver-scale beam experiments involving heavy-ion beams with I beam ∼1 A to provide guidance for the design of an Integrated Research Experiment (IRE) for driver system studies within the next 5 years. Multiple-beam sources and injectors are being designed and a one-beam module will be built and tested. Another experimental effort will be the transport of such a beam through ∼100 magnetic quadrupoles. The experiment will determine transport limits at high aperture fill factors, beam halo formation, and the influence on beam properties of secondary electron production and neutral atom desorption from the vacuum pipe wall. Research into driver technology will be briefly presented, including the development of ferromagnetic core materials, induction core pulsers, multiple-beam quadrupole arrays and plasma channel formation experiments for pinched transport in reactor chambers.

We simulate fusion power plant driver efficiency by pulsing small induction cores at 5 Hz (a typical projected power plant repetition rate), with a resistive load in the secondary winding that is scaled to simulate the beam loading for induction acceleration. Starting from a power plant driver design that is based on other constraints, we obtain the core mass and acceleration efficiency for several energy ranges of the driver accelerator and for three magnetic alloys. The resistor in the secondary is chosen to give the same acceleration efficiency, the ratio of beam energy gain to energy input to the core module (core plus acceleration gap), as was computed for the driver. The pulser consists of a capacitor switched by FETs, Field Effect Transistors, which are gated on for the desired pulse duration. The energy to the resistor is evaluated during the portion of the pulse that is adequately flat. We present data over a range of 0.6–5 μs pulse lengths. With 1 μs pulses, the acceleration efficiency at 5 Hz is measured to be 75%, 52%, and 32% for thin-tape-wound cores of nanocrystalline, amorphous, and 3% silicon steel materials respectively, including only core losses. The efficiency increases for shorter pulse durations.

We describe the goals and research program leading to the Heavy Ion Integrated Research Experiment (IRE). We review the basic constraints which lead to a design and give examples of parameters and capabilities of an IRE. We also show design tradeoffs generated by the systems code IBEAM.

The approach to heavy-ion-driven inertial fusion studied most extensively in the US uses induction modulators and cores to accelerate and confine the beam longitudinally. The intrinsic peak-current capabilities of induction machines, together with their flexible pulse formats, provide a suitable match to the high peak-power requirement of a heavy-ion fusion target. However, as in the RF case, where combinations of linacs, synchrotrons, and storage rings offer a number of choices to be examined in designing an optimal system, the induction approach also allows a number of architectures, from which choices must be made. We review the main classes of architecture for induction drivers that have been studied to date. The main choice of accelerator structure is that between the linac and the recirculator, the latter being composed of several rings. Hybrid designs are also possible. Other design questions include which focusing system (electric quadrupole, magnetic quadrupole, or solenoid) to use, whether or not to merge beams, and what number of beams to use – all of which must be answered as a function of ion energy throughout the machine. Also, the optimal charge state and mass must be chosen. These different architectures and beam parameters lead to different emittances and imply different constraints on the final focus. The advantages and uncertainties of these various architectures will be discussed.

The problem of storage and compression of intense bunches of heavy ions in the HIDIF driver accelerator is studied by means of a computer simulation with the particle-in-cell code PATRIC. It is shown that the requirements from final focusing (6 ns pulse length and Δ p / p =1%) can be reconciled with the linac output longitudinal emittance by careful optimization of the transfer into the barrier buckets.

The objective of this work is to minimize the cost of the materials and maximize the performance of magnetic cores, a major cost component of a Heavy-Ion-Fusion (HIF) induction accelerator driver. This includes selection of the alloy for cost and performance, and maximizing the performance of each alloy evaluated. The two major performance parameters are the magnetic flux swing and the energy loss. The volt seconds of the cores, obtained from the flux swing with Faraday's Law, determines the beam energy and duration. Core losses from forming domains and moving their boundaries are a major factor in determining the efficiency of an induction accelerator.

The crystal structures of three compositions of Na + β″-aluminogallate with different gallium content (Na 1+ x Mg x (Al 1− y Ga y ) 11− x O 17 , x =0.67; y =0.23, 0.30 and 0.41), have been studied by single-crystal X-ray diffraction (Mo Kα) 298 K. The β″-aluminogallates are isostructural with the β″-aluminas. The gallium atoms occupy predominantly tetrahedral sites in the spinel blocks. These substitutions cause increases in the lengths of the a - and c -axes. The Na + ion distribution in the conduction plane is influenced only slightly by increasing gallium substitution. The structural results are discussed in the light of earlier structural studies of Na + β″-alumina and Na + β″-gallate.

Transverse beam combining is a cost-saving option employed in many designs for induction linac heavy ion fusion drivers. The resultant transverse emittance increase, due predominantly to anharmonic space charge forces, must be kept minimal so that the beam remains focusable at the target. A prototype combining experiment has been built using the MBE-4 experimental apparatus. Four new sources produce up to 6.7 mA Cs+ beams at 200 keV. The ion sources are angled toward each other so that the beams converge. Focusing upstream of the merge consists of four quadrupoles and a final combined-function element (quadrupole and dipole). All lattice elements are electrostatic. Owing to the small distance between beams at the last element (about 3–4 mm), the electrodes here are a cage of small rods, each at different voltage. The beams emerge into the 30-period transport lattice of MBE-4 where emittance growth due to merging, as well as the subsequent evolution of the distribution function, can be diagnosed. The combiner design, simulation predictions and preliminary results from the experiment are presented.

We are currently constructing a prototype quadrupole electromagnet for the proposed Induction Linac Systems Experiments (ILSE) at LBNL. ILSE (PUB-5219, 1989 (Lawrence Berkeley Laboratory)) will address many physics and engineering issues relevant to the design of a heavy ion fusion driver accelerator. The pulsed electromagnet has two layers of current windings and will produce a field gradient of 28 T m −1 , with a usable aperture radius of 6 cm. It operates at a repetition rate of I Hz, steady state. In this paper, we discuss how the interaction of various concerns such as maximum dynamic aperture, short lattice period, field quality, iron yoke weight, heat transfer, and voltage stand-off have led to our particular design choices. We also present two- and three-dimensional numerical calculations concerning field topography and the results of transport simulations of space-charge-dominated ion beams with ILSE parameters.

A series of Na+β″-aluminogallate crystals with gallium substitution for aluminum ranging from 30 to 52 at% was grown by flux evaporation. X-ray diffraction analysis confirms that the structure is analogous to the rhombohedral Na+β″-alumina, and the a- and c-axis lattice parameters increase with increasing gallium content. Two-point ac impedance measurements indicate that the Na+β″-aluminogallates are fast ion conductors with conductivities equal to and slightly lower than that of Na+β″-alumina, 0.36 to 1.18 S/cm at 300°C, with comparable activation energies. Despite the increase in c-axis lattice parameter, the conductivities of the Na+ aluminogallates are actually somewhat lower as the c-parameter becomes larger. Conductivities of K+ and Ag+ ion exchanged crystals are also presented.

Longitudinal beam compression and the possibility of concurrent pulse shaping have been the subject of previous studies using analytical methods and 1D

A sixteen beam injector to supply 500mA per beam of C/sup +/ ions, which was initially designed and partially constructed at Los Alamos National Laboratory, is under further development at LBL. We report on development work involving single and multiple carbon arc sources. These sources are useful because they can easily supply the 25 mA/cm/sup 2/ needed for the device. They are pulsed sources which do not cause unacceptable background gas loads in the column. We present emittance and reproducibility data. Acceleration is by a high gradient column using a succession of aperture lenses to focus the beam. The electrodes are arranged with holes for parallel beams inside 28 inch alumina insulators. Data on beam propagation and electron trapping in the column is presented. The acceleration potential is provided by a Marx bank with inductive grading along the structure. The resulting 34 ..mu..sec rise time pulse allows the accelerating potential to equilibrate along the column before the current is injected. Acceleration occurs during a 1 ..mu..sec window at the peak of the pulse, kn which the voltage is flat within 0.1%. 8 refs., 7 figs.