TL;DR
This paper presents the first absolutely calibrated measurements of liquid xenon's response to MeV-scale gamma rays, crucial for next-generation neutrinoless double beta decay experiments. The study precisely determines the W-value (energy per detected quantum) and reveals deviations in recombination fluctuations at higher energies compared to lower-energy extrapolations, leading to an improved detector resolution model.
What Problem Does This Paper Address?
Detectors searching for rare events like neutrinoless double beta decay (0νββ) often utilize liquid xenon (LXe) due to its excellent scintillation and ionization properties. To accurately predict the sensitivity and performance of these detectors, especially future large-scale experiments like nEXO, a precise understanding of LXe's response to energy depositions is essential. This includes knowing the absolute yields of scintillation light and ionization charge produced by interactions, and how these yields change with electric fields. Previous measurements have primarily focused on lower energy depositions (keV scale), leaving a gap in our understanding at the MeV scale relevant for many calibration sources and potential background events. Furthermore, accurately modeling the energy resolution of these detectors requires a detailed understanding of recombination fluctuations, which are the anti-correlation between charge and light signals.
Methods
The researchers utilized data from the EXO-200 experiment, a single-phase LXe time projection chamber. They employed a suite of calibration sources ($^{228}$Th, $^{226}$Ra, and $^{60}$Co) emitting gamma rays in the 1-2.5 MeV range. A key aspect of their method was an "external charge-injection" calibration to precisely determine the gain of the charge readout electronics, achieving a high precision of 1.8% uncertainty. This allowed for an absolute calibration of the charge and light yields. They developed a semi-empirical model for the detector's energy resolution, incorporating measured detector parameters like electronics noise, recombination fluctuations, and photon detection efficiency. This model was then used to constrain the recombination efficiency, a crucial parameter for understanding the charge-light anti-correlation.
Datasets
The primary dataset used in this study is from the EXO-200 experiment, specifically calibration data collected during Phase I and Phase II operations. These datasets consist of recorded signals from the charge readout wires and the avalanche photodiodes (APDs) for various gamma-ray energies (2615 keV, 1764 keV, 1332 keV, and 1173 keV) at different applied electric fields. The paper also references and compares its findings with data from previous experiments and simulations, such as the Noble Element Simulation Technique (NEST) [9, 29].
Key Insights and Ideas
- MeV-Scale Response is Crucial: This work bridges a critical gap by providing the first absolutely calibrated measurements of LXe response at MeV energies, directly relevant for predicting the performance of future 0νββ detectors.
- Precise W-Value: The measured W-value of 11.5 ± 0.5 (syst.) ± 0.1 (stat.) eV is a fundamental parameter for energy scale calibration in LXe detectors.
- Recombination Deviations: Recombination fluctuations at MeV energies deviate from extrapolations of lower-energy data, highlighting the need for energy-dependent models.
- Improved Resolution Model: A semi-empirical model was developed that accurately describes the EXO-200's energy resolution across various electric fields, providing a valuable tool for detector design and analysis.
- Constrained Recombination Efficiency: The study successfully constrained the recombination efficiency ($\epsilon_r$), finding it to be very close to 1 (0.97), indicating that nearly all recombining electrons produce a detectable photon.
Key Results
- The W-value was measured to be $11.5 \pm 0.5$ (syst.) $\pm 0.1$ (stat.) eV.
- The photon detection efficiency ($\epsilon_p$) was determined to be $(8.1 \pm 0.5)\%$.
- The measured charge and light yields at MeV energies differ by approximately 10% and 20%, respectively, from NEST predictions.
- The measured energy resolution at 2615 keV and 567 V/cm was found to be $1.23\%$.
- The recombination efficiency ($\epsilon_r$) was constrained to $0.97$.
Limitations and Future Work
The authors acknowledge that the semi-empirical resolution model relies on the accuracy of measured detector parameters. While systematic errors have been carefully considered, unknown sources of systematic error or contributions to the overall resolution not included in the model could still influence the best-fit values, particularly for $\epsilon_r$. Future work could involve further refining the NEST simulation to better match these MeV-scale measurements, especially concerning the charge-to-light ratio differences observed. Additionally, more precise measurements of IR photon contributions to scintillation in LXe would be beneficial. The developed resolution model provides a strong foundation for predicting the performance of future LXe detectors, and further improvements could be achieved by reducing dominant noise sources, particularly in the photo-detector readout.
Summary updated