H. Vernon Smith

Status report on a dc 130 mA, 75 keV proton injector (invited)

A 110 mA, 75 keV dc proton injector is being developed at Los Alamos. A microwave proton source is coupled to a two solenoid, space-charge neutralized, low-energy beam transport (LEBT) system. The ion source produces 110 mA proton current at 75 keV using 600–800 W of 2.45 GHz discharge power. Typical proton fraction is 85%–90% of the total extracted ion current, and the rms normalized beam emittance after transport through a prototype 2.1 m LEBT is 0.20 (πmm mrad). Beam space-charge neutralization is measured to be >98% which enables the solenoid magnetic transport to successfully match the injector beam into a radio-frequency quadrupole. Beam simulations indicate small emittance growth in the proposed 2.8 m low-energy demonstration accelerator LEBT. The LEBT also contains beam diagnostics, steering, and a beam deflector for variable duty factor and accelerator fast protect functions. The injector beam availability status is also discussed.

A 75 keV, 140 mA proton injector

A dc and pulsed-mode 75 keV proton injector has been developed and is used in characterization of a continuous-wave 6.7 MeV, 100 mA radio-frequency quadrupole (RFQ). The injector is used frequently at the full RFQ design power (100 mA, 6.7 MeV) where the RFQ admittance (1 rms, normalized) is 0.23 (π mm mrad). The injector includes a 2.45 GHz microwave proton source and a beam space-charge-neutralized, two magnetic-solenoid, low-energy beam-transport system. The design RFQ beam transmission of 95% has been demonstrated at 100 mA RFQ output current.

The 4X Source

Our Penning surface-plasma source (SPS) discharge chamber was enlarged 4X in two dimensions. To date, three pulsed discharge modes have been studied: two with noisy arc (¿20% H- current fluctuations) and one with quiescent arc (¿1% H- current fluctuations). Lower arc magnetic field and higher H2 gas flow allow switching from the noisy to the quiescent mode. The noisy modes yield up to 120 mA of 29-keV H- beam; for 110 mA at 29 keV, the two-dimensional normalized rms emittance is 0.017 x 0.018 ¿·cm·mrad. The quiescent mode yields 75 mA of 29-keV H- beam; for 67 mA at 24 keV, the emittance is 0.011 x 0.012 ¿·cm·mrad.

Spectroscopic investigation of H− and D− ion source plasmas

Several H I (Balmer), Cs I, Cs II, and Mo I lines emitted by the small‐angle source and 4X source plasmas are studied. After correcting for Stark broadening, the Hα line width gives the H‐atom temeprature kTHO. After correcting for Doppler broadening, the Hβ and Hδ line widths give the electron density ne. For pulsed operation of both sources, kTHo is 1.5 to 2 eV and ne is 1 to 2×1014/cm3,with kTHo and ne scaling approximately with the square root of the discharge current. For the 4X source operated on D2, kTDo and ne are near the values of kTHo and ne obtained for H2 operation. Assuming that the H−/D− ion temperature equals the H/D‐atom temperature, we deduce a lower limit to the H−/D− beam emittance.

Penning surface-plasma source scaling laws---theory and practice

The small‐angle source (SAS), 4X source, and 8X source are Penning surface‐plasma sources (SPS) that produce high‐current, high‐brightness H− ion beams for accelerator applications. It is desirable to achieve high duty‐factor (df) operation, ultimately dc, with a Penning SPS. Two developments may make this goal possible. First, the H− beam‐emission scaling from the SAS (the 1X device) to the 4X source, and from the 4X source to the 8X source, is more favorable than the scaling laws predict. Second, fringe‐field separation of the e− and H− beams may make it possible to handle the power of the coextracted e− beam, especially since a collar arrangement reduces the e− loading. We compare our measured results with the predictions of the Penning SPS scaling laws. Particular attention is paid to the H− current and temperature scaling as well as the power efficiency.

H° temperature and density measurements in a Penning surface‐plasma H− ion source. I.

Using vacuum ultraviolet laser‐absorption spectroscopy, the H° density and temperature are measured as a function of discharge current and H2‐gas flow in both the plasma column and the drift region between the plasma column and the emitter in the 4X source. For typical source operating parameters, the atom temperature is 1.5 eV in the plasma column and 0.6 eV in the drift region; the atom density 7×1014 cm−3 in the plasma column and 4×1014 cm−3 in the drift region. Separate measurements give 2% for the ratio of H2 molecules in the first vibrational level to the total H2 density.

Electron suppression in the H− beam from a Penning surface‐plasma source

The ratio of electrons to negative ions extracted from Penning surface‐plasma sources such as the 8X source is low even before any further steps are taken to suppress the electrons. For the 8X source the e−/H− ratio is typically four or five to one for H− operation and nine to one for D− operation. Because the coextracted e− present a power‐loading problem to the 8X‐source extraction system, methods to dissipate and/or reduce the power in the e− beam must be developed before extracting a dc H− or D− beam. Thus, we systematically varied the geometry of a cylindrical collar installed in the near‐extraction region of the 8X source. The observed dependence of the extracted e− and H− currents on the collar radii and lengths suggests that a conical collar would provide superior electron suppression. The conical collar that we tested lowered the e−/H− ratio to 0.9/1 without reducing the extracted H− current.

Spatial distributions of the emitting species in a Penning surface‐plasma source

Using optical spectroscopy we study the spatial and temporal distributions of the Hα, Cs i(4555 A), Cs ii(4604 A), and Mo i(3903 A) emission lines in a Penning surface‐plasma source (SPS). A diagnostic slit exposes the entire SPS discharge gap either parallel or perpendicular to the magnetic field. The spatial and temporal distributions of the emitting species are recorded using a 1‐m monochromator. In addition, the visible light and the Hα and Cs ii(4604 A) spatial distributions are recorded with a video camera. The cesium atomic and ionic light, and the molybdenum atomic light, is strongly concentrated near the cathodes; the visible light and the Hα light is almost uniform in both directions. Electron‐impact ionization of atoms sputtered from the cathodes and the return of the ions to the cathodes by residual plasma fields is probably the mechanism which concentrates cesium near the cathodes. The Cs0 mean free path is estimated to be 16 and 0.43 mm for 2 and 400 A discharges, respectively.

H− temperature measurements by a slit diagnostic technique

H/sup -/ ion beams are extracted at 5-25 kV from a long, narrow slit on a Penning surface-plasma source (the 8X source). The extraction geometry produces negligible transverse electric field (focusing effects) along the slit length. Therefore, the ion angular spread reflects the distribution of ion energies at the plasma surface. The angular distributions are measured with an electric-sweep emittance scanner whose slits are oriented normal to the long dimension of the emission slit. The nearly Maxwellian angular distributions measured over the central portions of the ribbon beam give kT(H/sup -/) of 0.1 eV to 0.2 eV for a 2-A DC discharge and 0.8 eV to 1.0 eV for 350-A to 500-A pulsed discharges. This diagnostic technique has sufficient position resolution to allow measurement of the kT(H/sup -/) spatial distributions. It also allows study of the kT(H/sup -/) dependencies on ion source parameters (e.g. increasing the H/sub 2/ gas flow lowers kT(H/sup -/)).<<ETX>>

H- temperature dependences in a Penning surface-plasma source

Simple analysis of the nearly-Maxwellian angular distribution of the ribbon H{sup {minus}} ions beams extracted from a long, narrow slit on the 8X source yields the H{sup {minus}} temperature, kT{sub H{minus}}. The derived kT{sub H{minus}} are 0.1--0.3 eV for a 2-A dc discharge and 0.7-1.3 eV for a 400-A pulsed discharge. Because this diagnostic method relies on simple electronic techniques, it allows rapid study of the dependencies of kT{sub H{minus}} on the source parameters, such as gas flow and discharge current. These variations of kT{sub H{minus}} in the 8X source are qualitatively similar to those observed for the H-atom temperature, kT{sub H}o, in the 4X source, another Penning surface-plasma source. 10 refs., 4 figs.