Library

Notes: We have tried to shed more light on the possible mechanisms of the beneficial effect of gas mixing for highly charged ion production, in particular on the presence of an ion cooling effect. For the first time a method was applied to deduct the ion temperature from all measured ionic currents. The ion temperature values derived showed a clear decreasing trend in conjunction with mass changes of the gas mixture, consisting of "pure" argon (no mixing gas), argon plus natural oxygen, argon plus isotopic 18O, or argon plus 22Ne. Each of the applied mixing gases gives a higher charged ion (HCI) output (highest for 18O) as well as a lower ion current for the singly charged beam-particle output. The relative high particle fraction (about 40%) of singly charged nonbeam particles is an indication that effective plasma ion cooling is possible. Although the differences in ion temperature calculated are small, the effect is likely substantial, since the ion confinement has a dependency with (Ti)−5/2. Hydrogen currents (residual) are observed to be smallest for the best HCI-confined plasmas; so far no explanation for this can be given. © 2000 American Institute of Physics.

Notes: At the KVI there are presently four ion sources available: POLIS, for polarized protons and deuterons, to be injected in the new superconducting cyclotron AGOR ECRIS3, to produce highly charged ions for AGOR, ECRIS4, to produce highly charged ions for the Atomic and Surface Physics research facility (essentially a copy of ECRIS3), and CUSP, for high intensity AGOR beams of protons, deuteron, or alpha particles. We will report here on the performance and status of POLIS and ECRIS3. POLIS is an atomic beam type polarized ion source, according to the well known scheme with a cryo cooled dissociator, watercooled hexapole magnets, "strong field" and "weak field" radio frequency (rf) transitions for hydrogen and deuterium, and a 2.45 GHz electron cyclotron resonance (ECR) type ionizer. The source was constructed in a joint effort with the Forschungszentrum Karlsruhe, Cyclotron department. The ionizer section was modified and can be operated up to 40 kV. Following successful tests and deuteron polarization measurements at low energies, POLIS was connected to AGOR in November 1996. Most parts are now adapted to the AGOR-KVI control system. The first experiments with proton beams (190 MeV) from AGOR were done with polarized beams. As a consequence, the first measurement of proton polarization could be done only during these experiments, and all efforts to improve the polarization had to take place also during experiments. Efforts to get the weak field transition in operation have given a present best value of polarization of 70%±0.5% of the theoretical value. This value is deduced from an assumed analyzing power of the applied reaction. The strong field unit is not perfect; the best value which has been observed is 56%±0.5%. However, the difference between the given polarizations could be the result of a systematic error in the measurement of polarization "zero." The output of POLIS is in the order of 50 μÅ. The running time is usually 1–2 weeks (with nozzle at 53 K) until the dissociator and nozzle become contaminated; with a nitrogen layer (nozzle at 35 K) the polarization degree tends to be higher, but the running time is reduced to 3–5 days. POLIS has been in operation during more than 70% of the 33 weeks scheduled for experiments with AGOR since November 1996. Present activities concentrate on the improvement of the strong field unit, increasing the ionizer efficiency and the running time. ECRIS3 is a room temperature ECR ion source, with rf and gas feed according to the CAPRICE scheme. Its performance has been reported earlier (RIKEN 95). Gas mixing is important for the production of highly charged ions. Like in other sources, the anomaly for oxygen isotopes has also been measured. One new observation concerning gas mixing, in particular for the production of argon beams, is the following: For medium charge states, e.g., Ar8+, there is a slight preference for helium as a mixing gas, as compared to pure argon. In fact, oxygen is not good here. On the contrary, for high charge states like Ar14+, oxygen is a very good mixing gas, helium has only a small effect, but the best results have been obtained with isotopic 18O as a mixing gas. Recently, the same has been observed with CAPRICE 10 GHz at CEN-Grenoble, and to some extent with ECRIS4.© 1998 American Institute of Physics.

Notes: At the Workshop, the operation of various new and existing ECR ion sources was reported, with most of the emphasis on new methods to improve the performance and extend the variety of species. Much attention was paid to theoretical aspects, in particular to the basic question of electron heating; a complete investigation is still considered to be a long term project. A number of measurements of diagnostic character were presented. The strong radial component of the field in the magnetic trap turns out to play a crucial role in the confinement of hot electrons and hence in the production of very high charge state ion beams. A donor for cold electrons in any form—like a biased probe, plasma cathode, or electron gun—contributes strongly to increased output and stability of operation of the ECR ion source.

Notes: Some measurements have been done on the source ECRIS 1 at K.V.I., showing the beneficial effect to the output currents when a biased disk is positioned in the first stage, as reported by the Grenoble ECRIS group. The increase is in the order of 40% and it can be explained by the enrichment of the electron density in the plasma. Different saturation voltages for different charge states tend to confirm this explanation, suggesting new experiments, as, for example, the substitution of the first stage by an electron gun.

Notes: The performance of the KVI ECRIS-2 for a mixture of 18O, 17O, and 16O isotopes is−as expected−very much the same as for natural oxygen gas, i.e., the sum of the currents of given charge states of the isotopes equals that of the same charge state of the natural oxygen. However, it is found that the ratios of currents 18Oq+/17Oq+ and 18Oq+/16Oq+ increase with q. New measurements on ECRIS-1 (Minimafios) at KVI confirm this anomalous charge state distribution. This effect is also present when nitrogen or argon gas is mixed into the plasma. However, when applying helium as a mixing gas, the anomaly disappears. An explanation in terms of ion cooling (here transfer of kinetic energy to, in particular, the light species) seems valid.

Notes: The highly performing Electron Cyclotron Resonance Ion Source CAPRICE in Grenoble was operated with a mixture of three oxygen isotopes. The summed currents per charge state show a distribution almost identical to that of natural oxygen. However, the distributions per isotope are distinctly and systematically different in the sense that the heaviest isotopic distribution is most prominently peaked at high charge states. This anomalous effect reduces or disappears if helium is added to the isotopic mixture. The addition of either natural helium or isotopic 3He has no significantly differing result. Presently no quantitative model is available to satisfactorily explain the measurements. © 1996 American Institute of Physics.

Notes: The plasma confinement within an electron cyclotron resonance ion source significantly influences the charge state distribution and hence the performance of the source. The different axial and radial diffusion processes govern the confinement time. In many experiments it has been shown that negatively biasing the end plate in the injection region improves the charge state distribution. In a few x-ray and vacuum ultraviolet spectroscopy experiments to clarify the mechanism it is observed that the biasing improves the confinement of the plasma. It is estimated that the effect cannot be explained solely by secondary electron emission from the plate into the plasma. We propose that by biasing, the overall balance between radial ion losses and axial electron losses will change, resulting in a different diffusional mode of the entire plasma. Hence, the plasma potential and the average charge state of ions in the plasma are significantly influenced. Usually, the ion flux is dominating radial diffusion while the electron flux is dominating axial losses. This is possible due to compensating wall currents in the electrical conducting plasma chamber ("Simon short circuit"). Thus the usual approach of ambipolar diffusion does not hold in this situation. A similar effect takes place if the plasma chamber is coated with electrically insulating materials. The condition of overall flux balance to the walls is no longer fulfilled and has to be replaced by the local ambipolar particle movement. Again the entire diffusion profile of the plasma changes and the confinement improves. We examine the short circuit current as a measure for the diffusion mode in more detail and try to develop an approximate calculation on the influence of plasma potential and average charge ion state Z in the plasma. First results are presented and discussed. © 2002 American Institute of Physics.

Notes: This article deals with ion behavior in small open-ended magnetic devices, the electron cyclotron resonance ion sources (ECRIS) that were developed for multicharged ion production. The ECRIS are basically ECR-heated plasma confinement machines with hot electrons and cold ions. The main parameters of the ion population in ECRIS plasmas are successively analyzed: temperature, collisions, losses, confinement times, followed by the gas mixing effect, a specific technique to improve the performance as an ion source. A series of experiments is described for the systematic analysis of this effect. It is experimentally shown that high charge state optimization by gas mixing results from an ion confinement time improvement due to ion cooling, and relies on a compromise between three criteria, ion losses, mass effect, and ionization rates. © 1999 American Institute of Physics.

Notes: Abstract Elasticα-scattering angular distributions have been measured atE α=36.2 MeV, 39.6 MeV, 42.6 MeV, 49.5 MeV, 61.0 MeV for40Ca and atE α=36.2 MeV, 42.6 MeV, 49.5 MeV and 61.0 MeV for44Ca, respectively. At backward angles the data display an oscillatory structure forE α〈50 MeV and more smoothly decreasing slopes atE α =61.0 MeV resembling the data obtained at higher energies. The40Ca data belowE α =50 MeV show the well known backward enhancement, which at 42.6 MeV and 49.5 MeV can be fitted by aP 14 2 and aP 16 2 respectively. Together with previous data,P L 2-structures have now been observed for allL-values betweenL=8 andL=16. The slope of the curveL(L+1) versus excitation energy is slightly smaller forL〉12 than forL〈12. Optical model analyses (within a Woods-Saxon-folding model) lead to large differences between the44Ca and40Ca parameters. Furthermore, in our parametrization, the40Ca real potential depth shows dramatic changes with energy. This feature seems outside the domain of the optical model and requires consideration of additional effects (e.g. antisymmetrization) not included in the standard optical model. Present microscopic calculations on the basis of the Resonating Group Method are discussed in connection with the characteristic features ofα-scattering from40Ca.

Notes: Abstract K-shell ionisation probabilities and correspondingδ-electron spectra have been determined for 15 MeV protons and 50 MeV alpha particles on208Pb by measuring the energy spectrum of the scattered projectiles in coincidence with characteristic KX-rays, using a high resolution magnetic spectrograph. The results are in good agreement with SCA calculations.