James E. Stelzer

Effect of Ion Escape Velocity and Conversion Surface Material on H− Production

According to generally accepted models surface production of negative ions depends on ion escape velocity and work function of the surface. We have conducted an experimental study addressing the role of the ion escape velocity on H{sup -} production. A converter-type ion source at Los Alamos Neutron Science Center was employed for the experiment. The ion escape velocity was changed by varying the bias voltage of the converter electrode. It was observed that due to enhanced stripping of H{sup -} no direct gain of extracted beam current can be achieved by increasing the converter voltage. At the same time the conversion efficiency of H{sup -} was observed to vary with converter voltage and follow the existing theories in qualitative manner. We discuss the role of surface material on H{sup -} formation probability and present calculations predicting relative H{sup -} yields from different cesiated surfaces. These calculations are compared with experimental observations from different types of H{sup -} ion sources. The effects caused by varying cesium coverage are also discussed. Finally, we present a novel idea of utilizing materials exhibiting so-called negative electron affinity in H{sup -}/D{sup -} production under UV-light exposure.

H− Ion Source Development for the LANSCE Accelerator Systems

Employment of H‐ ion sources for the LANSCE accelerator systems goes back about 20 years, to the construction of the Proton Storage Ring (PSR). The standard ion source consists of a filament driven multi‐cusp discharge vessel with a biased converter electrode for negative‐ion production and an 80‐kV extraction system feeding into a 670‐kV electrostatic pre‐accelerator. The source typically delivers 18 mA pulsed beam current into the pre‐accelerator column and reaches up to 35 days between services at 60 Hz pulse repetition rate. Recent development efforts with this source have been dedicated to improved filament material, improved cesium oven geometry and operating the source at elevated temperatures. A second line of development focuses on filament‐less devices driven by a helicon discharge. Performance data obtained with the standard source as well as key results for the helicon experiments are given in this paper; the helicon work is discussed in a separate paper in much greater detail.

Tungsten filament material and cesium dynamic equilibrium effects on a surface converter ion source.

We present results on two different aspects that affect surface converter H(-) ion source performance: tungsten filament material and converter/wall temperature control. On the tungsten material aspect, evidence that filament grain size affects the source performance as well as filament failure modes is shown. Materials with impurity contents that hinder grain growth during conditioning or operation are to be avoided in order to increase the filament lifetime. Regarding the temperature control of the converter and plasma chamber walls, we present results of increased current output of up to 2.5 mA (15%). This is explained by generating increased cesium vapor pressure leading to enhanced sputtering of H(-) ions.

H/sup -/ surface converter source development at Los Alamos

Production of H/sup -/ ions by the surface conversion process is being pursued at the Los Alamos Neutron Science Center (LANSCE) as part of an upgrade project to provide higher currents and enhanced flexibility for injecting a 800 MeV H/sup -/ beam into the proton storage ring (PSR). An eventual goal of 40-mA H/sup -/ current at 80 keV beam energy with 0.13 /spl pi/mm-mrad (1 rms normalized) emittance at 12% duty factor (120 Hz, 1 ms) is desired. To attain this goal, two types of surface converter sources are being investigated in which H/sup -/ ions are extracted either radially or axially through a line-cusp magnetic field. The radial source produces 18 mA H/sup -/ in a 28-day run cycle for LANSCE production while the axial source has been developed to the desired 40 mA current. However, an emittance growth of factor 2-3 accompanied the increased axial source current. The axial source development program includes electron suppression, increased beam current, and 80-keV beam emittance measurements. The current understanding of the emittance growth mechanism will be discussed.

CW 8X ion source development

Using the 4X source performance and the Penning SPS scaling laws, we predicted the performance of the 8X source. A pulsed 8X source was then built and tested. After verifying the pulsed 8X source operation, especially the H− beam current, emittance, and power efficiency, we designed and built the CW 8X source. We plan to operate the source arc at dc power levels up to 30 kW. This will be accomplished by actively cooling the electrode surfaces with pressurized, hot water. The CW 8X source is presently undergoing shake‐down tests on a test stand at Los Alamos. Evaluation of the high‐current, hydrogen‐cesium dc arc will begin when these tests are completed.

A direct‐current Penning surface‐plasma source

After developing a pulsed 8X source for H− beams, we are now testing a cooled, dc version. The design dc power density on the cathode surface is 900 W/cm2, much higher than achieved in any previously reported Penning surface‐plasma source. The source is designed to accommodate dc arc power levels up to 30 kW by cooling the electrode surfaces with pressurized, hot water. After striking the arc using a 600‐V pulser, a 350‐V dc power supply is switched in to sustain the 100‐V discharge. Now our tests are concentrating on arc pulse lengths ≤1 s. Ultimately, the discharge will be operated dc. The source is described and the initial arc test results are presented.

Initial operation of the CW 8X H/sup -/ ion source discharge

A pulsed 8X source was built and the H/sup -/ beam current, emittance, and power efficiency were measured. These results were promising, so a cooled, dc version designed for operation at are power levels up to 30 kW was built. Testing of the CW 8X source discharge is underway. The design dc power loading on the cathode surface is 900 W/cm/sup 2/, considerably higher than achieved in any previous Penning surface-plasma source (SPS). Thus, the electrode surfaces are cooled with pressurized, hot water. We describe the source and present the initial operating experience and are test results.<<ETX>>

Helicon plasma generator-assisted surface conversion ion source for the production of H− ion beams at the Los Alamos Neutron Science Centera)

The converter-type negative ion source currently employed at the Los Alamos Neutron Science Center (LANSCE) is based on cesium enhanced surface production of H(-) ion beams in a filament-driven discharge. In this kind of an ion source the extracted H(-) beam current is limited by the achievable plasma density which depends primarily on the electron emission current from the filaments. The emission current can be increased by increasing the filament temperature but, unfortunately, this leads not only to shorter filament lifetime but also to an increase in metal evaporation from the filament, which deposits on the H(-) converter surface and degrades its performance. Therefore, we have started an ion source development project focused on replacing these thermionic cathodes (filaments) of the converter source by a helicon plasma generator capable of producing high-density hydrogen plasmas with low electron energy. In our studies which have so far shown that the plasma density of the surface conversion source can be increased significantly by exciting a helicon wave in the plasma, and we expect to improve the performance of the surface converter H(-) ion source in terms of beam brightness and time between services. The design of this new source and preliminary results are presented, along with a discussion of physical processes relevant for H(-) ion beam production with this novel design. Ultimately, we perceive this approach as an interim step towards our long-term goal, combining a helicon plasma generator with an SNS-type main discharge chamber, which will allow us to individually optimize the plasma properties of the plasma cathode (helicon) and H(-) production (main discharge) in order to further improve the brightness of extracted H(-) ion beams.

Evaluation of a simple method for chopping Penning surface‐plasma source H− beams

Accumulator rings proposed for use in high‐intensity spallation‐neutron sources require a chopped beam with ∼100‐ns‐wide particle‐free gaps at 1–2 MHz rates, with fall and rise times ≤20 ns. Chopping the beam directly in the ion source may be an attractive way to provide the desired beam structure. Previous measurements showed that placing a grounded collar in the drift region just before the emission aperture lowers the e−/H− ratio in the Penning surface‐plasma source (SPS) H− beam. We electrically isolated the collar and biased it to modulate the extracted H− current. Positive collar bias decreases the H− beam by up to 90%. The fastest H− current fall and rise times achieved to date are 400 ns and 2 μs, respectively. The current fall time is close to the 300‐ns pulser rise time. The current rise time is considerably longer than the 500‐ns pulser fall time. Negative collar bias lowers the H− beam by up to 50%. Simulations indicate that the beam time structure will be preserved in transport from the ion s...