Abstract
Searching for dark matter requires detectors with extremely low background radiation, as expected event rates are fewer than a few events per tonne-year. This research details the development of highly radiopure NaI(Tl) crystals, achieving concentrations of $^{nat}$K below 20 ppb and $^{210}$Pb below 5.7 $\mu$Bq/kg through optimized purification methods. We report on the design and performance of a prototype detector module intended for a large-scale search system.
The Challenge of Background Noise
The search for Dark Matter (DM) is primarily a quest to identify Weakly Interacting Massive Particles (WIMPs) amidst a "haystack" of radioactive background noise. Trace amounts of isotopes like Potassium-40 ($^{40}$K) and Lead-210 ($^{210}$Pb) within detector materials can mimic DM signals. While liquid Xenon detectors currently lead in sensitivity, Sodium Iodide (NaI(Tl)) crystals are essential for investigating spin-dependent interactions and verifying the controversial "annual modulation" signal claimed by the DAMA/LIBRA experiment.
Purification Methodology
To achieve the necessary radiopurity, a multi-stage protocol was developed:
- Double Recrystallization (RC×2): By exploiting solubility differences at varying temperatures, the team found that two rounds of recrystallization effectively reduced Potassium levels to the target threshold of < 20 ppb.
- Ion-Exchange Resins: Specialized resins were employed to target Lead ($^{210}$Pb) contamination, achieving a reduction factor of over 60.
- Pulse Shape Discrimination (PSD): This technique was used to distinguish between alpha particles (internal contamination) and the beta/gamma rays of interest. Alpha particles exhibit a shorter scintillation decay time (190 ns) compared to beta/gamma rays (230 ns), allowing for precise impurity quantification.
Experimental Results and Performance
The final test ingot (#85) achieved record-low impurity levels: $^{nat}$K at < 20 ppb and $^{210}$Pb activity at < 5.7 $\mu$Bq/kg. The prototype detector module demonstrated a low energy threshold of 1.6 keVee, which is critical for detecting low-energy WIMP recoils.
The background event rate near the threshold was measured at 1.5 day$^{-1}$kg$^{-1}$keV$^{-1}$. While current energy resolution is limited by the quantum efficiency of the photomultiplier tubes, future iterations aim to drop the background rate below 1.0.
Conclusion and Future Outlook
This work successfully demonstrates a scalable method for producing ultra-pure NaI(Tl) crystals. The next phase, PICOLON Phase-I, will deploy a four-module array of these crystals at the Kamioka Observatory in Japan. This experiment aims to provide a definitive, independent cross-check of the DAMA/LIBRA results and push the boundaries of spin-dependent dark matter searches.
Summary updated