⚛️ high-energy experiments

Scintillation characteristics of an undoped CsI crystal at low-temperature for dark matter search

This study demonstrates that undoped CsI crystals coupled with SiPMs exhibit significantly enhanced light output and decay times at low temperatures (86 K), establishing them as a promising, world-competitive detector for low-mass dark matter searches via spin-dependent interactions.

Original authors: W. K. Kim, H. Y. Lee, K. W. Kim, Y. J. Ko, J. A. Jeon, H. J. Kim, H. S. Lee

Published 2026-08-25

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: W. K. Kim, H. Y. Lee, K. W. Kim, Y. J. Ko, J. A. Jeon, H. J. Kim, H. S. Lee

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Deep beneath the Earth's surface, in the quiet silence of underground laboratories, scientists are listening for a whisper that has never been heard. They are searching for dark matter, an invisible substance that makes up most of the universe's mass but refuses to interact with light or ordinary matter in any way we can easily detect. To find it, researchers build sensitive detectors that wait for a rare, tiny collision between a dark matter particle and an atom in their equipment. The challenge is that these collisions are incredibly faint, often happening at energies so low that the signal is easily drowned out by the background noise of the environment. To hear this whisper, scientists need materials that glow brightly when struck, even by the smallest hit, and they need to listen in a setting where the temperature is cold enough to silence the thermal chatter of the atoms themselves.

In this pursuit, a team of researchers turned their attention to a specific type of crystal: undoped cesium iodide. Unlike the crystals often used in medical scanners or other detectors, which are mixed with other chemicals to make them glow, this material is pure. The scientists wanted to see if cooling this pure crystal down to the freezing point of liquid nitrogen would make it a better listener. They attached two tiny, highly sensitive light sensors directly to a small, one-gram piece of the crystal and placed the entire assembly inside a vacuum chamber. By carefully controlling the temperature, they could watch how the crystal behaved as it cooled from room temperature down to 86 Kelvin, a temperature just above absolute zero.

What they found was a dramatic transformation. As the crystal grew colder, it began to glow much more brightly when hit by radiation. At room temperature, the crystal produced a modest amount of light, but as the temperature dropped to 86 Kelvin, the light output increased more than twelve times. This surge in brightness is crucial because it means the detector can see much smaller events that would otherwise be invisible. Along with the brighter glow, the crystal also changed how it responded over time. The flash of light it emitted lasted much longer in the cold, stretching from a few dozen nanoseconds at room temperature to nearly six hundred nanoseconds at the lowest temperature. This slower decay gives the detector more time to analyze the shape of the signal, which helps scientists distinguish between a genuine dark matter hit and random background noise.

The researchers also paid close attention to the sensors themselves, which are made of silicon and are known to generate their own internal noise, especially when warm. They measured how often these sensors fired falsely and found that this noise dropped to almost nothing as the temperature fell. By carefully accounting for these factors, they confirmed that the bright, slow, and quiet behavior of the cold crystal was real and reliable. They used this small-scale test to imagine a much larger experiment. They proposed building a detector using 200 kilograms of this same undoped crystal, arranged in large blocks and read out by arrays of these silicon sensors.

Because the atoms in cesium iodide have a specific property related to their internal spin, this material is particularly good at hunting for a specific type of dark matter that interacts with protons. The team ran simulations to see how well this 200-kilogram detector would perform if it were cooled to liquid nitrogen temperatures. The results suggest that such a machine would be incredibly sensitive, capable of detecting dark matter particles with masses between 60 million and 2 billion electron volts. In this specific range of particle weights, the proposed detector could reach a level of sensitivity that rivals or surpasses the best experiments currently operating in the world. The study concludes that by combining the unique properties of pure cesium iodide with modern silicon sensors and the power of extreme cold, scientists have a promising new tool to explore the hidden corners of the universe, potentially opening a window into a type of dark matter that has remained out of reach until now.

Technical Summary: Scintillation Characteristics of Undoped CsI at Low Temperatures for Dark Matter Search

Problem Statement
The search for rare events, such as dark matter interactions and coherent elastic neutrino-nucleus scattering (CE $\nu$ NS), requires detectors with low energy thresholds. In scintillating crystals, lower thresholds are strongly associated with higher light yields. While doped crystals (e.g., CsI(Tl)) are commonly used at room temperature, previous studies suggest that undoped CsI and NaI crystals exhibit significantly higher light yields at cryogenic temperatures. However, a comprehensive characterization of undoped CsI coupled directly with Silicon Photomultipliers (SiPMs) across a broad temperature range, specifically for the purpose of optimizing future dark matter detectors, was needed. Additionally, the potential of undoped CsI for probing dark matter-proton spin-dependent interactions requires evaluation, given that both Cesium and Iodine possess odd numbers of protons.

Methodology
The study utilized a 1.0 g undoped CsI crystal (5.8 mm $\times$ 5.9 mm $\times$ 7.0 mm) directly coupled to two Hamamatsu SiPMs (model S13360-2172) without optical grease or pads to avoid uncertainties at low temperatures. The assembly was housed in a vacuum cryostat equipped with a temperature control unit, allowing measurements from room temperature (293 K) down to 86 K.

Key experimental procedures included:

Calibration: A $^{109}$ Cd radioactive source was used to generate 22.1 keV, 25.0 keV, and 88.0 keV peaks. Due to energy resolution limitations, the x-ray peaks were treated as a combined 23.0 keV peak for analysis.
Noise Characterization: The Dark Count Rate (DCR) and Crosstalk (both internal and external) were measured across various overvoltages (3 V to 8 V) and temperatures. Corrections for these noise sources were applied to the light yield calculations.
Measurements: The team measured light yield, energy resolution, and scintillation decay times at various temperatures. The SiPM gain was stabilized by maintaining a constant overvoltage (specifically 5 V for most temperature-dependent analyses) to ensure consistency.
Simulation: To evaluate future sensitivity, the authors simulated a 200 kg experiment using 8.7 kg modules arranged in a 5 $\times$ 5 array. They incorporated the Migdal effect to assess sensitivity to low-mass WIMP-proton spin-dependent interactions, utilizing ionization probabilities from iodine and xenon as proxies for cesium.

Key Results

Light Yield: A significant increase in light yield was observed as temperature decreased. At 86 K, the light yield reached 26.2 $\pm$ 1.3 photoelectrons (PE) per keV (electron-equivalent) at 8 V overvoltage. This represents a 12-fold increase compared to room temperature performance.
Decay Time: The scintillation decay time increased substantially with decreasing temperature, ranging from 76.0 $\pm$ 6.0 ns at room temperature to 605.2 $\pm$ 3.2 ns at 86 K.
Energy Resolution: Improved energy resolution was observed at lower temperatures, correlating with the enhanced light output.
Noise Performance: The DCR decreased dramatically at lower temperatures (e.g., dropping from ~700 kHz at room temperature to 0.2 kHz at 150 K). However, data below 150 K were excluded from the analysis because the DCR dropped further, resulting in poor statistical reliability. Crosstalk was found to be highly dependent on bias voltage but independent of temperature.

Significance and Claims
The authors claim that coupling undoped CsI crystals with SiPMs at low temperatures creates a promising detector for rare event searches, particularly for low-mass dark matter. The study posits that the combination of high light yield, the odd proton numbers of Cs and I (making them sensitive targets for spin-dependent interactions), and the low background capabilities of existing crystal growth technology (utilizing purified powder from the KIMS experiment) offers a competitive advantage.

Specifically, the paper evaluates a hypothetical 200 kg undoped CsI experiment operating at 77 K with a 1-year exposure and a background rate of 1 count/kg/keV/day. Under these conditions, the proposed detector is projected to achieve world-competitive sensitivity for WIMP-proton spin-dependent interactions in the mass range of 60 MeV/c $^2$ to 2 GeV/c $^2$ . This sensitivity is achieved by leveraging the Migdal effect to lower the effective energy threshold to approximately 0.08 keV (5 PE threshold). The authors conclude that this approach allows for the exploration of a new dark matter parameter space that is difficult to access with current technologies.

More like this