First $\beta -\nu$ correlation measurement from the recoil-energy spectrum of Penning trapped ${}^{35}\mathrm{Ar}$ ions

We demonstrate a novel method to search for physics beyond the standard model by determining the $\beta -\nu$ angular correlation from the recoil-ion energy distribution after $\beta$ decay of ions stored in a Penning trap. This recoil-ion energy distribution is measured with a retardation spectrometer. The unique combination of the spectrometer with a Penning trap provides a number of advantages, e.g., a high recoil-ion count rate and low sensitivity to the initial position and velocity distribution of the ions and completely different sources of systematic errors compared to other state-of-the-art experiments. Results of a first measurement with the isotope ${}^{35}\mathrm{Ar}$ are presented. Although currently at limited precision, we show that a statistical precision of about 0.5% is achievable with this unique method, thereby opening up the possibility of contributing to state-of-the-art searches for exotic currents in weak interactions.

Figure 1

Scheme of the WITCH Penning traps and the spectrometer with an example trajectory of a recoil ion from the traps to the detector. For a figure of the complete setup see, e.g., Ref. [23].

Figure 2

Overview of the two data sets obtained in this experiment. Data set 1 (bottom, dashed curve) is the sum of all cycles with no retardation barrier applied. Data set 2 (top, solid curve) is the sum of all cycles with retardation voltages applied in some of the time bins (inset zoomed). The argon ions are kept in the cooler trap from $t=0$ s until $t=0.5$ s and in the decay trap from $t=0.5$ s until $t=2$ s.

Figure 3

(a) PHD of particles arriving on the MCP detector, without retardation voltage, i.e., with recoil ions ($t=0.55$ s), and with 600 V ($t=0.60$ s) being applied, which blocks all recoil ions. (b) Rebinned difference in counts for the two PHDs in the upper panel with statistical error bars and a Gaussian fit (${\chi }^2/\nu =0.7)$, which is typical for ions.

Figure 4

Difference in counts between data set 2 (obtained with retardation voltages applied) and the scaled data set 1 (without retardation voltages applied). Raw data and the data corrected for the ${}^{35}\mathrm{Ar}$ half-life and losses in the decay trap; see text for details.

Figure 5

Recoil-ion energy spectrum for ${}^{35}\mathrm{Cl}$ daughter ions. The black line corresponds to the best fit to the data, yielding $a=1.12\pm 0.{33}_{\text{stat}}$. For comparison the curves simulated for $a=+1$ (pure vector interaction) and $a=-1$ (pure scalar interaction) are also shown. The SM value is $a=0.900\left(2\right)$ [25]. Lines are drawn to guide the eye.