Calibration of liquid argon and neon detectors with ${}^{83}\mathrm{Kr}$${}^m$

We report results from tests of ${}^{83}\mathrm{Kr}$${}^m$ as a calibration source in liquid argon and liquid neon. ${}^{83}\mathrm{Kr}$${}^m$ atoms are produced in the decay of ${}^{83}\mathrm{Rb}$, and a clear ${}^{83}\mathrm{Kr}$${}^m$ scintillation peak at $41.5$ keV appears in both liquids when filling our detector through zeolite coated with ${}^{83}\mathrm{Rb}$. Based on this scintillation peak, we observe 6.0 photoelectrons/keV in liquid argon with a resolution of $8.2\%$ ($\sigma /E$) and $3.0$ photoelectrons/keV in liquid neon with a resolution of $19\%$ ($\sigma /E$). The observed peak intensity subsequently decays with the ${}^{83}\mathrm{Kr}$${}^m$ half-life after stopping the fill, and we find evidence that the spatial location of ${}^{83}\mathrm{Kr}$${}^m$ atoms in the chamber can be resolved. ${}^{83}\mathrm{Kr}$${}^m$ will be a useful calibration source for liquid argon, neon dark matter, and solar neutrino detectors.

Figure 1

Energy levels in keV of ${}^{83}\mathrm{Rb}$. For $75\%$ of the time, ${}^{83}\mathrm{Rb}$ decays to ${}^{83}\mathrm{Kr}$${}^m$ , which in turn decays via electromagnetic transitions with energies of 32.1 and 9.4 keV. The half-life of ${}^{83}\mathrm{Kr}$${}^m$ to decay via the first transition to the $7/2^{+}$ state is 1.83 h and the half-life for the subsequent decay to the stable $9/2^{+}$ state of ${}^{83}\mathrm{Kr}$ is 154 ns.

Figure 2

Schematic representation of the scintillation cell.

Figure 3

The gas-handling system for the ${}^{83}\mathrm{Kr}$${}^m$ runs. In active mode, valve V1 is closed while valves V2 and V3 are open. In passive mode, valves V1 and V3 are open while valve V2 is closed.

Figure 4

Example of a ${}^{83}\mathrm{Kr}$${}^m$ event in liquid argon from a single PMT, digitized by the 12-bit CAEN WFD, sampling at 250 MHz. The 32.1- and 9.4-keV components of this particular ${}^{83}\mathrm{Kr}$${}^m$ event are highlighted.

Figure 5

Energy spectrum of ${}^{83}\mathrm{Kr}$${}^m$ runs in argon, with (bottom) and without (top) a background subtraction. The light yield is 6.0 photoelectrons/keV and the resolution is $8.2\%$ ($\sigma /E$) at 41.5 keV.

Figure 6

Light yield vs energy in argon, referenced to the value of 6.0 pe/keV measured for the ${}^{83}\mathrm{Kr}$${}^m$ peak. There is a $1\%$ systematic error on each point stemming from variations in the position of the ${}^{83}\mathrm{Kr}$${}^m$ peak from run to run.

Figure 7

Rate of ${}^{83}\mathrm{Kr}$${}^m$ events in argon as a function of time from the beginning of a fill. The rate reaches a steady state at 170 ${\mathrm{s}}^{-1}$.

Figure 8

Rate of ${}^{83}\mathrm{Kr}$${}^m$ events in argon as a function of time from ending a fill. The rate decays with a fitted half-life of $1.82\pm 0.0$ h, consistent with the reported value of $1.83\pm 0.02$ h.

Figure 9

(Top) Rate of ${}^{83}\mathrm{Kr}$${}^m$ events in argon as a function of time from the beginning of the fill. (Bottom) The mean value of the asymmetry parameter $A$ in argon during the same fill. The ${}^{83}\mathrm{Kr}$${}^m$ initially appears at the top of the detector, temporarily raising $A$, before filling the whole active region as discussed in the text.

Figure 10

Rate of ${}^{83}\mathrm{Kr}$${}^m$ events in argon as a function of time from the beginning of the fill when running in passive mode. In this mode, the rate of ${}^{83}\mathrm{Kr}$${}^m$ is $10\%$ of that observed for the active mode.

Figure 11

Energy spectrum of a background-subtracted ${}^{83}\mathrm{Kr}$${}^m$ run in neon. As discussed in the text, the extrapolated light yield is $3.0\pm 0.3$ pe/keV, and the resolution is $19\%$ ($\sigma /E$) at 41.5 keV.

Figure 12

Rate in arbitrary units of ${}^{83}\mathrm{Kr}$${}^m$ in neon as a function of time from the end of a fill. The rate decays with a fitted half-life of $1.16\pm 0.56$ h, consistent with the value in the literature.