Antihydrogen Production within a Penning-Ioffe Trap

Slow antihydrogen ($\overline{\mathrm{H}}$) is produced within a Penning trap that is located within a quadrupole Ioffe trap, the latter intended to ultimately confine extremely cold, ground-state $\overline{\mathrm{H}}$ atoms. Observed $\overline{\mathrm{H}}$ atoms in this configuration resolve a debate about whether positrons and antiprotons can be brought together to form atoms within the divergent magnetic fields of a quadrupole Ioffe trap. The number of detected $\overline{\mathrm{H}}$ atoms actually increases when a 400 mK Ioffe trap is turned on.

Figure 1

Outside (a) and cross section (b) views of cylindrical ring electrodes that are biased to form Penning traps for $\overline{\mathrm{p}}$, ${\mathrm{e}}^{+}$, and ${\mathrm{e}}^{-}$. In the direction of the central axis (vertical in the experiment) is a 1-T bias field from a solenoid (not shown). The axial and radial fields of a quadrupole Ioffe trap come from two pinch coils and four racetrack coils, respectively.

Figure 2

Cylindrical ring electrodes (black) with well depth contours for ground-state $\overline{\mathrm{H}}$ atoms in the Ioffe trap (a). Potential wells seen by charges on-axis (b) during (red) and after (green) $\overline{\mathrm{H}}$ production, with corresponding equipotentials for these two cases (d)–(f). An $\overline{\mathrm{H}}$ can field ionize and deposit its $\overline{\mathrm{p}}$ in the detection well if it travels this direction after production in the nested well. Magnetic field lines (blue) that are parallel to the trap axis with no radial Ioffe field (e) are significantly redirected (c) when the Ioffe field is added (d),(f).

Figure 3

Illustration of the sharp radial cutoff outside of which no charged particle can be trapped (a), and the dependence of this cutoff upon Ioffe trap depth (b). This cutoff shows up as a sharp edge in the spectrum of $\overline{\mathrm{p}}$ released (c) from the green well of Fig. 2.

Figure 4

Number of $\overline{\mathrm{H}}$ ionized in the detection well per trial within the Ioffe trap (a) and within only the axial field of the Ioffe trap (b), as a function of the trap depth. $\overline{\mathrm{H}}$ numbers are normalized to an average of 0.1 million $\overline{\mathrm{p}}$ per $\overline{\mathrm{H}}$ production and detection trial, with reproducibility error bars.