CDEX-1B Experiment Overview

• The CDEX-1B experiment leverages a 1kg pPCGe detector with a 0.88 mm dead layer to achieve an ultra-low 160 eVee threshold for rare event detection.
• It implements rigorous background suppression through layered shielding, anti-Compton vetoes, and advanced rise-time discrimination techniques.
• The study extends searches to sub-GeV WIMPs and axion-like particles, setting benchmarks for future ton-scale dark matter experiments.

The CDEX-1B experiment is a direct dark matter search conducted at the China Jinping Underground Laboratory (CJPL) using a 1 kg-scale p-type point-contact high-purity germanium detector. This experiment represents an evolved phase of the China Dark Matter Experiment (CDEX) series, aiming to achieve ultra-low background and sub-200 eV electron-equivalent thresholds for sensitive detection of rare events—including low-mass weakly interacting massive particles (WIMPs), axion-like particles, and neutrinoless double beta decay. The CDEX-1B program demonstrates significant advances in noise reduction, background modeling, and analysis methodology, setting benchmarks for future ton-scale experiments.

1. Experimental Design and Detector Configuration

The CDEX-1B employs a p-type point-contact germanium (pPCGe) detector with a total mass of 1008 g and an effective fiducial mass of 939 g after accounting for a lithium-diffused surface dead layer of approximately 0.88 ± 0.12 mm (Zhang et al., 2023, Liang et al., 9 Oct 2025). The central p+ point-contact electrode (∼1 pF capacitance) enables a low electronic noise environment, which is critical for achieving physics analysis thresholds as low as 160 eV electron-equivalent energy (eVee) (Yang et al., 2017, Wang et al., 2019). The assembly is shielded by layered oxygen-free copper, borated polyethylene, and lead. An active NaI(Tl) anti-Compton veto system surrounds the detector, with nitrogen flushing employed to mitigate radon backgrounds (Ma et al., 2019).

Key hardware innovations realize:

Feature	Design/Performance	References
Point-contact pPCGe detector	1 kg-class, 0.88 mm dead layer	(Yang et al., 2017)
Energy threshold	160 eVee (with ∼17% signal efficiency)	(Yang et al., 2017, Wang et al., 2019)
Fiducial mass	939 g	(Liang et al., 9 Oct 2025, Zhang et al., 2023)
Anti-Compton system	NaI(Tl) scintillator, active veto	(Ma et al., 2019)
Environmental mitigation	Nitrogen flushing, multi-layer passive	(Ma et al., 2019)

Pulse acquisition utilizes multi-shaping (6 μs, 12 μs) and timing amplifier outputs, digitized by 100 MHz, 14-bit FADCs. Selection cuts—including bulk/surface discrimination via rise-time analysis, anti-coincidence with veto signals, and PSD—are applied to suppress backgrounds and isolate potential rare-event signals (Jiang et al., 2018, Li et al., 2022).

2. Background Modeling, Calibration, and Data Analysis

CDEX-1B’s background model integrates detailed simulations and experimental constraints. Gamma-ray backgrounds from the laboratory environment are characterized by in situ measurements and Monte Carlo (Geant4) simulations, revealing uranium-thorium-potassium concentrations in CJPL concrete walls of 6.8 ± 1.5 Bq/kg (²³⁸U), 5.4 ± 0.6 Bq/kg (²³²Th), and 81.9 ± 14.3 Bq/kg (⁴⁰K) (Ma et al., 2020). Cosmogenic activation in the Ge detector, notably from ⁶⁸Ge, ⁶⁵Zn, and ³H, is evaluated using the formalism

$$R_i = \sum_j N_j \int \Phi_k(E) \sigma_{ijk}(E)\, dE$$

where $N_j$ is the atomic abundance, $\Phi_k(E)$ the cosmic ray flux for each species, and $\sigma_{ijk}(E)$ the production cross section (Ma et al., 2018, Nie et al., 2023). Validation is achieved by comparing observed cosmogenic peaks (e.g., 10.37 keV from ⁶⁸Ge) with simulated spectra.

Bulk/surface event separation employs rise-time ($\tau$) analysis with PDFs constructed for both populations. Signal and background probabilities are extracted via maximum likelihood and, for rare-event searches, by profile likelihood ratio or Markov Chain Monte Carlo sampling (Wang et al., 2019, Liang et al., 9 Oct 2025). Energy calibration is performed using both internal (cosmogenic) and external sources, with <1% linearity deviation and sub-220 eV FWHM at 10 keVee for bulk events (Jiang et al., 2018, Wang et al., 2019).

A model to treat and remove anomalous fast bulk events (FBEs), arising from the region near the passivation layer, is implemented by fitting the rise-time distribution using PDFs obtained from combined simulation and calibration data. This technique exploits the detector’s effective single-hit spatial resolution, enabling additional background discrimination (Li et al., 2022, Jiang et al., 2018).

3. Dark Matter Searches: Elastic, Inelastic, and Sub-GeV Channels

CDEX-1B sets competitive constraints on both spin-independent (SI) and spin-dependent (SD) WIMP-nucleon interactions. The lowered threshold (160 eVee) allows sensitivity down to WIMP masses of 2 GeV (Yang et al., 2017, Ma et al., 2019). For SI nuclear recoils, the exclusion limit is quoted as $\sigma^{SI}_{\chi N} \sim 2\times10^{-40}$ cm² at $m_\chi \simeq 3$ GeV (Yang et al., 2017).

By incorporating the Migdal effect—the emission of atomic electrons during nuclear recoil—CDEX-1B extends its search to sub-GeV WIMPs, probing masses down to 50–75 MeV/$c^2$ with upper limits on $\sigma^{SI}_{\chi N}$ of $2\times10^{-32}$ to $7\times10^{-35}$ cm² in the time-integrated analysis and $3\times10^{-32}$ to $9\times10^{-38}$ cm² in the annual modulation channel (Liu et al., 2019, Ma et al., 2019). The relevant detection rate formula is

$$d^2R / (dE_{EM} dE_R) = N_T \frac{\rho_\chi}{m_\chi} \int v f_v(\vec v + \vec v_E) \bigg( \frac{d^2\sigma}{dE_{EM} dE_R} \bigg) d^3v$$

with $E_{EM}$ the electromagnetic energy from Migdal ionization and $E_R$ the nuclear recoil energy.

CDEX-1B further reports on inelastic dark matter (iDM) scenarios, where WIMP-nucleus scattering excites the WIMP to a higher state $\chi^*$ split by an energy $\delta$. The minimum velocity required is

$$v_\mathrm{min} = \frac{1}{\sqrt{2 E_{nr} m_N}}\left(\frac{m_N E_{nr}}{\mu} + \delta\right)$$

where $m_N$ is the nucleus mass, $\mu$ the WIMP-nucleus reduced mass, and $E_{nr}$ the nuclear recoil energy. Detailed Geant4 background simulations and Bayesian MCMC parameter fitting yield 90% C.L. SI cross-section exclusion curves that rule out DAMA/LIBRA allowed parameter regions for $m_\chi = 250$–$500$ GeV and $\delta$ up to 50 keV (Liang et al., 9 Oct 2025).

Constraints from a nonrelativistic effective field theory (NREFT) approach are presented for 14 operator classes, enhancing sensitivity to non-SI/SD couplings (Wang et al., 2020).

4. Rare Event Physics Beyond WIMPs

CDEX-1B is utilized for axion and axion-like particle (ALP) searches. With 737.1 kg-day exposure at 160 eV threshold, the experiment constrains the axion–electron coupling $g_{Ae} < 2.48\times10^{-11}$ (CBRD channels: Compton, bremsstrahlung, atomic-recombination, de-excitation), and the product $g^{\mathrm{eff}}_{AN} g_{Ae} < 4.14\times10^{-17}$ from the ¹⁴.⁴ keV $^{57}$Fe solar axion channel, at 90% C.L. (Wang et al., 2019). Constraints for ALPs and vector bosonic dark matter via the axio-electric effect are also established.

A dedicated search for solar axions via the Bragg-Primakoff effect exploits the crystal periodicity of germanium, integrating over reciprocal lattice vectors,

$$\frac{dR}{dE} \propto \sum_{\vec G} \frac{d\Phi}{dE} \frac{|S(G)|^2}{|G|^2} \frac{d\sigma}{d\Omega} \delta\left(E - \frac{\hbar c |G|^2}{2 \hat{k} \cdot G}\right)$$

yielding a 95% C.L. upper limit $g_{A\gamma} < 2.08\times10^{-9}$ GeV$^{-1}$ for $m_A < 100$ eV/$c^2$ and excluding KSVZ hadronic axion masses above 5.3 eV/$c^2$ (Yang et al., 12 May 2024).

Neutrinoless double beta decay ($0\nu\beta\beta$) studies exploit the detector’s intrinsic energy resolution and background suppression. After 504.3 kg-day exposure, the observed background in the 1989–2089 keV ROI is 0.33 counts/(keV kg yr), yielding $T_{1/2}^{0\nu} > 1.0\times10^{23}$ yr (90% C.L.), corresponding to limits $\langle m_{\beta\beta}\rangle < 3.2$–$7.5$ eV (Zhang et al., 2023). This demonstrates CDEX-1B’s dual capability for both dark matter and $0\nu\beta\beta$ physics.

5. Background Mitigation and Cosmic Ray Activation Management

Cosmogenic activation—especially of ⁶⁸Ge, ³H, and related nuclides—constitutes a major source of background, both at sub-keV and MeV scales. The activation rates are quantitatively modeled using measured cosmic-ray fluxes, cross sections, and the detailed fabrication history (including altitude corrections via $\Phi_k(H) = \Phi_k(0)\exp[(p(H)-p(0))/\lambda_k]$). The total background rate in the $0\nu\beta\beta$ ROI, after 1 year underground cooling, is dominated (99%) by ⁶⁸Ge, with $49.6$ cpkty estimated for tonne-scale detectors (Nie et al., 2023, Ma et al., 2018).

Mitigation strategies include:

• Shielding during transport (low-carbon steel, thick polyethylene neutron shields) to reduce ⁶⁸Ge activation by up to an order of magnitude.
• Storage underground during non-working hours (reducing daily effective exposure) and allowing extended cooling before deployment ($\sim$2 years reduces ⁶⁸Ge by >60%).
• Material selection and purification for cryostats, support structures, and electronics (Nie et al., 2023).

6. Technological Evolution and Future Prospects

CDEX-1B serves as a reference for subsequent experimental expansion. The CDEX-10 prototype (∼10 kg, liquid nitrogen-immersed array) demonstrates scalability, maintaining sub-keV thresholds and consistent bulk event spectra compared to CDEX-1B, while achieving a background of 2 counts/(keV kg day) in the 2–4 keV window (Jiang et al., 2018, Ma et al., 2019). CDEX-50 and CDEX-1T (up to a ton-scale liquid nitrogen-cooled HPGe arrays) are planned to achieve background reductions by $\sim 10^{3}$ and further four orders of magnitude improvement in sensitivity, respectively (Liang et al., 9 Oct 2025, Nie et al., 2023).

In parallel, in situ $\gamma$-background characterization in CJPL-II’s Hall-C and extensive simulation frameworks (SAGE, Geant4) support the design and operation of these large-scale arrays (Ma et al., 2020).

7. Impact and Context in Rare Event Physics

CDEX-1B provides strong new limits on low-mass WIMP interactions (both elastic and inelastic), sub-GeV dark matter, axion couplings, and $0\nu\beta\beta$ decay, often setting the leading bounds in the relevant mass and coupling regions for high-purity germanium detectors. The experiment demonstrates that with ultralow-threshold, actively shielded pPCGe detectors and advanced discrimination techniques—including fast bulk event removal via rise-time modeling—rare event searches at millikelvin cross-section levels and background indices $\sim 0.3$ counts/(keV kg yr) are achievable in deep underground laboratories.

By establishing and validating methods for background suppression, energy calibration, cosmogenic activation control, and data analysis, CDEX-1B sets the technical foundation for next-generation experiments that will probe deeper into the parameter space of dark matter, axion-like particles, and lepton-number-violating processes.