Evaporative Cooling of Antiprotons to Cryogenic Temperatures

We report the application of evaporative cooling to clouds of trapped antiprotons, resulting in plasmas with measured temperature as low as 9 K. We have modeled the evaporation process for charged particles using appropriate rate equations. Good agreement between experiment and theory is observed, permitting prediction of cooling efficiency in future experiments. The technique opens up new possibilities for cooling of trapped ions and is of particular interest in antiproton physics, where a precise $CPT$ test on trapped antihydrogen is a long-standing goal.

Figure 1

(a) Simplified schematic of the Penning-Malmberg trap used to confine the antiprotons and of the two diagnostic devices used. The direction of the magnetic field is indicated by the arrow. (b) Potential wells used to confine the antiprotons during the evaporative cooling ramp. The antiprotons are indicated at the bottom of the potential well (red), and the different wells are labeled by their on-axis depth.

Figure 2

The number of antiprotons lost from the well as its depth is reduced is integrated over time and plotted against the well depth. The well depth is ramped from high to low; thus, time flows from right to left in the figure. The measured number is corrected for the 25% detection efficiency. The curves are labeled in decreasing order of the temperatures extracted from an exponential fit, shown as the solid lines. The temperatures (corrected as described in the text) are: A: 1040, B: 325, C: 57, D: 23, E: 19, and F: 9 K. As the antiprotons get colder, fewer can be used to determine their temperature, an effect described in Ref. [13].

Figure 3

(a) Temperature vs the on-axis well depth. The error is the combined statistical uncertainty from the temperature fit and an uncertainty associated with the applied potentials (one $\sigma$). The model calculation described in the text is shown as a line. (b) The fraction of antiprotons remaining after evaporative cooling vs on-axis well depth. The uncertainty on each point is propagated from the counting error (one $\sigma$). The initial number of antiprotons was approximately 45 000  for an on-axis well depth of $(1484\pm 14)\mathrm{mV}$.

Figure 4

The evaporation time scale vs applied on-axis well depth for five experiments with different ramp times. The on-axis well depth at the end of the ramp is $(10\pm 4)\mathrm{mV}$ in all cases. ${\gamma }^{-1}$ is indicated by the dot-dashed line, model calculations are shown as lines and, dashed lines indicate $\eta <4$ in the model calculation.