⚛️ nuclear experiments

Development of a 10 mol% Rubidium-doped CsI Crystal for $^{87}$ Rb Beta-Spectroscopy and Sterile Neutrino Searches

This paper reports the development and characterization of a novel 10 mol% rubidium-doped CsI scintillator crystal, which utilizes intrinsic $^{87}$ Rb beta-decay to enable high-efficiency source-in-detector spectroscopy for investigating forbidden beta-decay mechanisms and searching for keV-scale sterile neutrinos.

Original authors: W. K. Kim, K. W. Kim, L. T. Truc, H. S. Lee, H. J. Kim, Y. D. Kim

Published 2026-08-18

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: W. K. Kim, K. W. Kim, L. T. Truc, H. S. Lee, H. J. Kim, Y. D. Kim

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 within the subatomic world, certain atoms are unstable and seek stability by transforming into other elements, a process known as radioactive decay. One such atom, rubidium-87, is a quiet but persistent traveler that slowly turns into strontium-87 by releasing a tiny, fast-moving electron. This specific transformation is unusual because it is "forbidden" by the simplest rules of physics, meaning it happens very slowly and follows a complex path that scientists have studied for decades to understand the fundamental forces inside the nucleus. Beyond just mapping this atomic change, the way these electrons are released holds a potential key to a much larger mystery: the existence of invisible particles called sterile neutrinos. These hypothetical particles are thought to be a form of dark matter, and if they exist, they would leave a tiny, tell-tale mark on the energy of the electrons, like a subtle kink in a smooth curve. To find this mark, researchers need a detector that can catch these electrons with extreme precision, without losing any of their energy along the way.

A team of researchers has developed a new type of crystal detector designed specifically to solve this problem. They grew a large, clear crystal made of cesium iodide, a material that flashes with light when struck by radiation, and they deliberately mixed a significant amount of rubidium into the crystal itself. By embedding the radioactive rubidium directly inside the crystal, the scientists created a setup where the source of the electrons and the detector are one and the same. This "source-in-detector" approach ensures that the low-energy electrons from the decay are captured immediately, losing no energy to the air or container walls before they can be measured. The team grew these crystals using a method that slowly pulls molten material through a temperature gradient, allowing the atoms to arrange themselves into a solid, coherent block. They tested two versions: one with just rubidium and another with a tiny amount of thallium added alongside the rubidium to see if it improved the crystal's performance.

The results revealed a fascinating difference between the two materials. The crystal containing only rubidium behaved in a split manner, producing two distinct types of light flashes. Some flashes were bright and slow, showing that the rubidium was working as intended, while others were dim and fast, behaving exactly like pure cesium iodide without the rubidium. This suggested that the rubidium was not perfectly mixed throughout the crystal, leaving some areas that did not participate in the desired reaction. However, when the researchers added the small amount of thallium, the crystal changed completely. It produced a single, uniform type of light flash that was much brighter and more consistent. The thallium acted like a stabilizer, ensuring that the entire crystal responded in the same way, which is crucial for getting an accurate measurement of the electron's energy. The new crystal with thallium was found to be more than three times more efficient at converting the energy of the electrons into detectable light than the crystal with rubidium alone.

Using this improved crystal, the team measured the energy spectrum of the electrons emitted by the rubidium over a five-hour period while the detector was submerged in a liquid designed to block outside interference. They carefully analyzed the shape of the data to understand the complex physics of the decay and to look for the signature of a heavy sterile neutrino. Their measurements confirmed the expected behavior of the decay, providing precise numbers that describe how the electrons are distributed in energy. When they specifically searched for the "kink" that would indicate a heavy sterile neutrino, they found no such distortion in the data. Based on this lack of evidence, they calculated that if such a particle exists with a specific mass, its presence in the mix of neutrinos must be extremely small, far below the level they could detect with this experiment. While this result does not prove that sterile neutrinos do not exist, it sets a strict limit on how much they could be mixed with the known neutrinos. The researchers plan to repeat this experiment with a much larger detector deep underground to gather more data and reduce background noise, hoping to either find the elusive particle or rule it out with even greater certainty.

Technical Summary: Development of a 10 mol% Rubidium-doped CsI Crystal for 87Rb Beta-Spectroscopy and Sterile Neutrino Searches

Problem Statement
The third-forbidden non-unique beta-decay of Rubidium-87 ( $^{87}\text{Rb}$ ) to Strontium-87 ( $^{87}\text{Sr}$ ) serves as a critical benchmark for understanding forbidden beta-decay mechanisms and constraining nuclear matrix elements (NMEs) and the effective axial-vector coupling constant ( $g_A$ ). Furthermore, precision measurements of the $^{87}\text{Rb}$ beta-spectrum offer a pathway to search for physics beyond the Standard Model, specifically keV-scale sterile neutrinos, which would manifest as a kink-like distortion in the spectrum. A primary challenge in this field is achieving high detection efficiency for low-energy electrons while minimizing energy loss. Traditional external source configurations suffer from self-absorption and scattering; therefore, an "source-in-detector" configuration, where the radioactive isotope is intrinsic to the scintillator, is desirable. However, incorporating a high concentration of Rubidium into a Cesium Iodide (CsI) crystal presents significant material science challenges regarding crystal coherence and the stability of the scintillation response.

Methodology
The authors developed and characterized novel CsI scintillators doped with a nominal 10 mol% Rubidium Iodide (RbI) using the Bridgman growth method. Two distinct crystal configurations were investigated:

CsI:Rb: A crystal grown with 10 mol% RbI loading.
CsI:Tl,Rb: A crystal grown with 10 mol% RbI and co-doped with 0.1 mol% Thallium Iodide (TlI) to act as a co-dopant.

The crystals were grown from purified CsI powder (originating from the KIMS experiment) and high-purity RbI/TlI powders in a nitrogen-purged glovebox to maintain water levels below 300–400 ppm. The growth process utilized a two-zone vertical Bridgman furnace with a temperature gradient of approximately 10 °C/cm and a translation speed of 0.6 mm/h.

For characterization, the crystals were coupled to Hamamatsu H11934-300-10 extended green bialkali photomultiplier tubes (PMTs) and housed in aluminum enclosures wrapped in PTFE for light reflection and moisture protection. Data acquisition involved a 500 MHz, 12-bit flash-analog-to-digital converter (FADC) triggered by coincident pulses from dual PMTs. Scintillation properties, including light yield, energy resolution, and pulse shape characteristics, were evaluated using external gamma sources ( $^{241}\text{Am}$ , $^{133}\text{Ba}$ , $^{57}\text{Co}$ , $^{137}\text{Cs}$ ). The $^{87}\text{Rb}$ beta-spectrum was subsequently measured with the detector immersed in a linear-alkylbenzene (LAB) liquid scintillator to shield against external backgrounds.

Key Contributions and Results

Crystal Growth and Scintillation Populations: The study successfully grew coherent CsI ingots with 10 mol% Rb loading. A critical finding was the observation of two distinct scintillation populations in the pure CsI:Rb crystal. Pulse shape analysis revealed a "Type-I" branch (higher charge, longer decay time) consistent with Rb doping, and a "Type-II" branch (lower charge, shorter decay time) resembling pure CsI. This suggests non-uniform distribution of the Rb dopant or incomplete activation. In contrast, the CsI:Tl,Rb crystal exhibited a single, well-defined population, indicating that Tl co-doping effectively suppresses the pure CsI-like component and stabilizes the scintillation response.
Light Yield and Resolution: The light yield of the CsI:Rb crystal (Type-I events) was measured at $1.38 \pm 0.01$ PE/keV. The CsI:Tl,Rb crystal demonstrated a significantly higher light yield of $4.73 \pm 0.13$ PE/keV (a factor of 3.4 improvement). The energy resolution for the CsI:Tl,Rb crystal was parameterized using gamma lines from 59.54 keV to 661.66 keV, yielding a fit of $\sigma/E = p_0/\sqrt{E} + p_1$ with $p_0 = 60.19$ and $p_1 = 2.42$ .
Non-proportionality: The relative photon response of the CsI:Tl,Rb crystal was found to broadly follow the characteristic response of conventional CsI:Tl, with a slight increase in non-proportionality at few-keV energies. A quadratic function was fitted to this data to parameterize the non-proportionality for subsequent beta-spectrum analysis.
Beta-Spectroscopy and Shape Factors: Using the CsI:Tl,Rb crystal, the authors measured the $^{87}\text{Rb}$ beta-spectrum over a 5-hour exposure. By fitting the spectrum with a function incorporating the detector response, Fermi function, and a constant background, they extracted the shape factor coefficients for the third-forbidden non-unique decay: $a = 0.3481 \pm 0.0018$ and $b = 0.00616 \pm 0.00007$ . These values are consistent with previous experimental results and theoretical calculations.
Sterile Neutrino Search: Utilizing the measured spectrum and fixed shape factor coefficients, the authors searched for a sterile neutrino admixture. For a representative sterile neutrino mass of $m_4 = 106.3$ keV/c $^2$ , a statistical-only upper limit of $|U_{e4}|^2 < 6.8 \times 10^{-3}$ was established at the 95% confidence level.

Significance and Claims
The paper claims that the development of a 10 mol% Rb-doped CsI crystal provides a viable "source-in-detector" configuration for $^{87}\text{Rb}$ beta-spectroscopy, offering high detection efficiency and minimal energy loss for low-energy electrons. The study demonstrates that while high-loading Rb doping introduces complex scintillation behaviors (dual populations), the addition of a Tl co-dopant resolves these issues, yielding a stable, high-light-yield detector suitable for precision spectroscopy.

The authors present a preliminary basis for future sterile neutrino searches using this material. They explicitly note that the reported limit on sterile neutrino mixing is derived from statistical uncertainties only, with systematic uncertainties (energy calibration, non-proportionality, background modeling) remaining to be fully evaluated. The paper concludes that with higher statistics and improved shielding (planned at Yemilab), this approach holds promise for precise measurements of NMEs, $g_A$ , and sensitive searches for sterile neutrino admixtures.

More like this