Lossless Positron Injection into a Magnetic Dipole Trap

Editors' Suggestion

Lossless Positron Injection into a Magnetic Dipole Trap

E. V. Stenson, S. Nißl, U. Hergenhahn, J. Horn-Stanja, M. Singer, H. Saitoh, T. Sunn Pedersen, J. R. Danielson, M. R. Stoneking, M. Dickmann, and C. Hugenschmidt

Phys. Rev. Lett. 121, 235005 – Published 5 December 2018

Abstract

The high-efficiency injection of a low-energy positron beam into the confinement volume of a magnetic dipole has been demonstrated experimentally. This was accomplished by tailoring the three-dimensional guiding-center drift orbits of positrons via optimization of electrostatic potentials applied to electrodes at the edge of the trap, thereby producing localized and essentially lossless cross-field particle transport by means of the $E \times B$ drift. The experimental findings are reproduced and elucidated by numerical simulations, enabling a comprehensive understanding of the process. These results answer key questions and establish methods for use in upcoming experiments to create an electron-positron plasma in a levitated dipole device.

Received 21 May 2018

DOI:https://doi.org/10.1103/PhysRevLett.121.235005

Physics Subject Headings (PhySH)

Laboratory studies of space & astrophysical plasmas Plasma transport

Electron-positron plasmas Magnetized plasma

Atom & ion trapping & guiding

Plasma Physics

Authors & Affiliations

E. V. Stenson^1,2,3,*, S. Nißl^1,2, U. Hergenhahn^1,4, J. Horn-Stanja¹, M. Singer², H. Saitoh^1,5, T. Sunn Pedersen^1,6, J. R. Danielson³, M. R. Stoneking⁷, M. Dickmann², and C. Hugenschmidt²

¹Max Planck Institute for Plasma Physics, 17491 Greifswald, Germany and 85748 Garching, Germany
²Technische Universität München, 85748 Garching, Germany
³University of California, San Diego, La Jolla, California 92093, USA
⁴Leibniz Institute of Surface Engineering (IOM), 04318 Leipzig, Germany
⁵The University of Tokyo, 277-8561 Kashiwa, Japan
⁶University of Greifswald, 17489 Greifswald, Germany
⁷Lawrence University, Appleton, Wisconsin 54911, USA

^*evs@ipp.mpg.de

Confinement of Positrons Exceeding 1 s in a Supported Magnetic Dipole Trap

J. Horn-Stanja, S. Nißl, U. Hergenhahn, T. Sunn Pedersen, H. Saitoh, E. V. Stenson, M. Dickmann, C. Hugenschmidt, M. Singer, M. R. Stoneking, and J. R. Danielson

Phys. Rev. Lett. 121, 235003 (2018)

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 121, Iss. 23 — 7 December 2018

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) The experiment setup. (b) A typical simulation of lossless $E \times B$ transport from field lines connecting to the beam line (top), across wall-limited magnetic field lines (including the separatrix, magenta), and into the confinement region. Points along the trajectories of 100 particles—launched from the same starting point with a realistic range of velocities and propagated for $0.6 μ s$ —are projected onto the $x z$ plane (cyan dots), as is a single trajectory with values of $E_{∥}$ and $E_{⊥}$ near the median of the distribution (red line).
Reuse & Permissions
Figure 2
2D scans such as these in $I_{r} V_{E \times B}$ were used to experimentally optimize injection conditions into the dipole trap. When biases are applied to electrodes rw1 and top1, and the rest of the wall was grounded (a), high-efficiency injection—up to 100%—is achieved over a large region of the parameter space; this is not possible when all wall electrodes are biased (b). Single-particle simulations have been used to generate a “synthetic diagnostic” that produces good agreement with experiments (c),(d). (a)–(d) The color scale indicates the fraction of the beam that reaches the target probe.
Reuse & Permissions
Figure 3
Spatial profile measurements are shown for lossless injection and for injection with the constraint that the entire outer wall be set to the same bias. Counts detected by the scintillator ( $\times$ and $+$ , left axis; “probe out” value subtracted) and positron flux (solid symbols, right axis; measured with the charge-integrating amplifier) are each plotted as a function of target probe position and are in good agreement, as are synthetic profile measurements from simulations (open symbols). The black line and gray area indicate the average and standard deviation of measurements of total beam flux immediately upstream of the experiment; these and the vertical scatter of identical symbols illustrate typical uncertainties.
Reuse & Permissions
Figure 4
Other configurations for injection—e.g., (a) grounding rw1 (while leaving the magnet and top1 biased) or (b) grounding the entire trap (except the $E \times B$ plates)—are possible, but are less efficient and much more localized in parameter space. As in Fig. 2, the synthetic diagnostic (c),(d) produces good qualitative and quantitative agreement.
Reuse & Permissions
Figure 5
Toroidal movement in the dipole trap was simulated for various cases of interest. (a) Lossless injection using localized biases. The entire trajectory for a representative particle is shown in blue (the first $0.6 μ s$ ) and black (thereafter); midplane crossings for the entire beam are shown in red, projected onto the bottom plane. (b) Midplane crossings for the entire beam, for conditions with the entire wall at 5.5 V and the magnet at $- 15 V$ . (c) Two particles in an ideal dipole trap (without electrostatic potentials or beam line magnetic fields), started with the same velocity vector but different radii.
Reuse & Permissions
Figure 6
The islands that are seen in the 2D optimization scans of the $V_{top 1 & 2} V_{E \times B}$ parameter space plane when wall segment rw1 is grounded (a)—in contrast to when rw1 is biased above the beam energy (b)—were reproduced by the synthetic diagnostic (c). Examination of particle trajectories (d) shows that each island corresponds to particles entering the confinement region after a different number of electrostatic reflections off the biased upper wall segment. (The three different $V_{top 1 & 2}$ and $V_{E \times B}$ settings used for the trajectory simulations shown in (d)—labeled 0, 1, and 2 for the number of reflections—are indicated by asterisks in (c).)
Reuse & Permissions

Physical Review Letters