To fill the gap in domestic experimental data on neutron radiative capture reactions, the Key Laboratory of Nuclear Data at China Institute of Atomic Energy has developed a gamma-ray total absorption facility (GTAF), which consists of 40 BaF₂ detection units. Excluding the two units for neutron beam entrance and exit, the BaF₂ crystals form a spherical shell with an outer radius of 25 cm and an inner radius of 10 cm, covering approximately 95.2% of the solid angle and enabling nearly full-absorption measurements of γ-rays. In 2019, the facility was relocated to the back-streaming white neutron source of the China Spallation Neutron Source (CSNS), where research on neutron radiative capture cross section measurement methods was carried out.

After capturing a neutron, the nuclide forms an excited compound nucleus. The GTAF measures the cascade γ-rays emitted during the de-excitation of the compound nucleus, records (n, γ) reaction events, and determines the energy of the incident neutrons causing the reaction using the time-of-flight method. Comprehensive analysis of multiple experimental measurements on ¹⁹⁷Au sample verified the long-term stability of the energy resolution of the BaF₂ detector units and established the experimental conditions for online measurements: a coincidence time window of 64 ns, a multiplicity of M_γ ≥ 2, a summed γ-ray energy threshold of 3.5 MeV ≤ E_sum ≤ 8 MeV, and a standard neutron beam spot size of \phi 30 mm. These conditions were adopted to reduce backgrounds from electronic noise, natural environmental radiation, and (n, γ) reactions in other materials.

To further suppress backgrounds caused by scattered neutrons at the sample position, the vacuum pipe and a matching neutron absorber made of boron-loaded polyethylene were designed and optimized. Quantitative analysis of neutron sensitivity, together with comparisons to experimental data obtained under air conditions, verified the effectiveness of the vacuum pipe and neutron absorber configuration. The neutron sensitivity of the measured results remained nearly constant as the accelerator beam power increased. It was expected that the future upgrade of the CSNS Phase-II accelerator to 500 kW will not significantly affect the signal-to-background ratio of online experiments. Given the sample size of \phi 40 mm, measurements confirmed that using the standard \phi 30 mm neutron beam spot results in low neutron sensitivity. Under two experimental conditions-measurement in air and measurement with an aluminum-alloy vacuum pipe and a matching neutron absorber-several background components in the ¹⁹⁷Au experimental data were subtracted separately. Using the saturated resonance peak normalization method combined with the Back-n neutron energy spectrum, preliminary results of the (n, γ) reaction yield spectra and cross sections were obtained. Comparisons show that the vacuum pipe and a matching neutron absorber significantly improve the signal-to-background ratio, yielding experimental results closer to evaluated data; this configuration will therefore be adopted in subsequent experiments.

At present, the experimental cross sections of ¹⁹⁷Au measured by the GTAF are in good agreement with the Evaluated Nuclear Data File (ENDF) in the resolved resonance region, verifying the reliability of the facility and measurement method. Future plans to optimize the data processing include: correcting the detection efficiency ε for different incident neutron energies by combining Monte Carlo simulations with experimental measurements; reducing uncertainties induced by the double-bunch neutron beam using unfolding algorithms, applying corrections for multiple scattering effects, and analyzing resonance parameters of the ¹⁹⁷Au(n, γ) reaction with the SAMMY code to obtain more precise cross sections; conducting a quantitative uncertainty analysis component by component to determine the total uncertainty of the ¹⁹⁷Au(n, γ) cross section data.

The GTAF offers the advantage of nearly 4π solid-angle coverage and full γ-ray absorption, making it well-suited for measuring (n, γ) cross sections of small samples and nuclides with small cross sections—especially fissile nuclides such as ²³⁵U and ²³⁹Pu. It facilitates the use of anti-coincidence techniques to subtract backgrounds from fission reactions, providing important experimental data support for the development of the domestic nuclear industry.