Nonadiabatic electron gun at an electron beam ion source: Commissioning results and charge breeding investigations

The electron gun of the REXEBIS charge breeder at the REX/HIE-ISOLDE facility at CERN has been upgraded from a standard magneto-immersed type to a gun using a nonadiabatic magnetic element. The results from the cathode emission and electron beam propagation tests are presented, as well as the charge breeding efficiency for the new design. Complete mass-scans of the extracted beam have been performed from which the level of cathode-originating contaminations could be established, as well as partial pressures of the most abundant residual gases in the ion trapping region. Furthermore, optimal breeding times for a broad range of elements and charge states, either introduced as a gas or externally injected as singly charged ions into the trapping region, are given for different electron currents. From these values, effective electron current densities have been derived. Finally, the axial ion energy distributions of various elements and charge states were also measured, and the derived ion temperatures were correlated with the ion and electron beam overlap factors.

Figure 1

Left: design of REXEBIS electron gun with an IrCe cathode and iron ring as NA element. Right: enlarged view showing the cathode unit and Wehnelt.

Figure 2

Electron beam envelopes for beam currents of 300 mA, 500 mA, and 800 mA. The magnetic flux density at the cathode is $\approx 700\mathrm{Gs}$ and the purple curve shows its value along the $z$-axis. Solid lines represent the beam envelope for a fixed potential difference between the cathode and the post anode of 7.8 kV, while the dotted lines are for an adjusted potential difference between the cathode and the post anode (10.8 kV for 0.3 A and 6.8 kV for 0.8 A). NB: The vertical dimensions of the machine geometry are not to scale.

Figure 3

Dependence of the relative electron beam oscillation on the shift of the electron gun and iron ring from the optimum position for: (blue) a fixed potential on the post anode (PA) ($U_{\text{post anode}}=0\mathrm{V}$); (orange) optimized post anode potential. The post anode potential for the optimized case is presented with the black curve.

Figure 4

Envelopes of a 700 mA electron beam generated with the NA electron gun positioned at different magnetic field strengths. The shown magnetic field distribution (black), with the dip at the axial position of the NA element, has been optimized for a cathode positioned for a cathode flux density of 700 Gs.

Figure 5

Dependence of simulated electron beam current density versus magnetic field on the cathode surface of the NA gun. The electron current is 0.7 A and the magnetic field at the end of the simulation model is 1.73 T. Tracking simulation results are marked with black squares (TRAK) and triangles (CST). The corresponding current densities predicted by the Herrmann theory are indicated with a dashed red line. The blue dots mark the extrapolated current density inside the full magnetic field of 2 T.

Figure 6

Photograph of the NA electron gun assembly with the threefold symmetric support and intermediate plate to the left, and to the right the isolating ceramic adapter piece aligning the gun to the first drift tube.

Figure 7

IrCe cathode mounted inside the Wehnelt body. Left: x-ray tomography picture of the assembly showing the cathode base with four connection rods providing the current to the cathode heating, the set screws for radial positioning, as well as Wehnelt body and cup. Right: photograph of the same assembly, but with the Wehnelt cup removed.

Figure 8

Schematic view of REXEBIS with the NA electron gun installed. The axial magnetic field strength and potentials on the electrodes are indicated. All voltages are given with respect to the EBIS platform potential. The dimensions are not to scale.

Figure 9

Layout of the REX low-energy stage. The transfer lines are fully electrostatic and the $A/Q$-separation takes place in the separator magnet between FC3 and FC4. With a kicker element at the top of the BTS section, the extracted ion beam can be sent to the separator section.

Figure 10

Measured emission current as a function of cathode-to-anode voltage, with fitted experimental perveance value.

Figure 11

Measured losses as a function of the collector current. For collector currents above 300 mA, the voltage settings were adjusted to minimize the losses.

Figure 12

EBIS global efficiency as a function of breeding time for different elements. Systematic errors are not indicated.

Figure 13

EBIS single charge state efficiencies as a function of average charge state for different elements and an electron beam current of 200 mA.

Figure 14

For a selection of elements, the EBIS single charge state efficiency values have been divided by EBIS global efficiency. As a comparison to these curves, simulated relative abundances for the different charge states using the operational electron beam parameters are indicated as bars.

Figure 15

EBIS single charge state efficiency versus charge state for Na charge bred with 200 mA and 300 mA electron beams.

Figure 16

EBIS single charge state efficiencies for ${}^{152}\mathrm{Sm}$ injected into the EBIS trapping region with different beam energies (given in the legend). The vertical lines mark the mean and standard deviation of the charge state spectrum for the three injection energies.

Figure 17

Plot of the charge state evolution of ${}^{23}\mathrm{Na}$ with a beam current of 300 mA. The data points in the upper plot show the measured ion intensity (normalized to the primary beam accumulation time). The banded curves represent the evolution of the charge state distribution and its uncertainty according to the fitted theoretical model. The lower plot displays the fitted current density values at the sampling times. The uncertainty of the fitted parameters and the uncertainty bands of the theoretical evolution have been estimated with a Monte Carlo technique (see main text).

Figure 18

Collective plot of the fitted current density values for all experiments. The upper plot contains the current densities as obtained from the fitting model. The lower plot shows the corresponding effective radii, obtained assuming a top hat density profile of the electron beam and ion cloud. The orange lines display the theoretical electron beam radii calculated for different magnetic field configurations at the cathode.

Figure 19

Plot of the deduced electron current densities (at 200 mA electron current) for various elements versus the element number. The markers and error bars represent the arithmetic mean and standard deviation of the corresponding current density values shown in Fig. 18.

Figure 20

Plot of the charge state evolution of ${}^{39}\mathrm{K}$ with a beam current of 200 mA. The upper plot contains the measured ion output. The lower plot shows a simulation of the charge state evolution for an assumed current density of $200\mathrm{A}/{\mathrm{cm}}^2$.

Figure 21

Examples of axial energy measurements for ${}^{152}{\mathrm{Sm}}^{44+}$ recorded at a beam current of 300 mA. The data is overlaid with fitted energy distributions (see main text for explanation). The breeding times and fitted temperatures for the three cases are given in the legend.

Figure 22

Plot of the axial energy spectrum fit parameters for ${}^{152}\mathrm{Sm}$ (300 mA beam current). Both the temperature (upper graph) and the axial energy offset (lower graph) have been normalized by the ion charge under investigation (indicated in the legend) to yield equivalent voltages.

Figure 23

Plot of the axial energy spectrum fit parameters for ${}^{23}\mathrm{Na}$ (200 mA beam current). See Fig. 22 for further explanation.

Figure 24

Comparison plot of the temperature evolutions in various charge breeding scenarios (top). The element and electron beam current are indicated in the legend. Every data point represents several charge states; the error bars show the standard deviation of the temperature across all sampled charge states at a given breeding time. In the lower plot the temperature is rescaled with the charge state and the depth of the space charge potential well within the electron beam.

Figure 25

Neutralization and extracted charge per pulse as a function of charge breeding time for a 202 mA, 6200 eV electron beam. The recorded pressures in the gun and collector vacuum crosses were $1.2\times {10}^{-10}\mathrm{mbar}/1.2\times {10}^{-10}\mathrm{mbar}$ and $7.2\times {10}^{-10}\mathrm{mbar}/7.7\times {10}^{-10}\mathrm{mbar}$, respectively, with the valve to REXTRAP being closed/open.

Figure 26

The upper figure shows mass-scans acquired using a Faraday cup while sweeping the magnetic field of the $A/Q$-separator. Isotopes with relative abundances less than or equal to 1% are not labeled. The lower figure shows the dataset for $T_{\text{breed}}=10\mathrm{ms}$ with a logarithmic scale and identification of ${}^{13}\mathrm{C}$ (abundance 1.07%) and ${}^{15}\mathrm{N}$ (0.36%).

Figure 27

Particle energy histograms for $A/Q$-selections of 4.21 (top) and 4.244 (bottom) with detected ${}^{140}\mathrm{Ce}$ and ${}^{193}\mathrm{Ir}$ peaks together with other contaminants at similar mass-to-charge ratios.

Figure 28

Plot of the evolution of the mean and standard deviation (error bar) of the charge state distribution of various elements for electron beam currents of 200 mA and 300 mA.