We present improved constraints on couplings of solar axions and more generic bosonic dark matter particles using 737.1 kg-days of data from the CDEX-1B experiment. The CDEX-1B experiment, located at the China Jinping Underground Laboratory, primarily aims at the direct detection of WIMPs using a p-type point-contact germanium detector. We develop the profile likelihood ratio method for analysis of data in the presence of backgrounds. The background modeling is compatible with the data and no excess signals are observed. An energy threshold of 160 eV was achieved. This significantly improve the sensitivity for the bosonic dark matter below 0.8 keV. Limits are also placed on the coupling gAe < 2.26 × 10 −11 from Compton, bremsstrahlung, atomic-recombination and deexcitation channels and g ef f AN × gAe < 4.14 × 10 −17 from a 57 Fe M1 transition at 90% confidence level. All the constrains improve over our previous results.

In this paper, the on-line monitoring technology of the slow positron beam intensity has been applied to measure the Ag L αβγ characteristic X-ray yields with pure thick target by 6-9 keV positrons impact. The experimental characteristic X-ray yields have been compared with the corresponding yields simulated by Monte Carlo calculations, and the ratio of the former to the latter has been defined as the scaling factor. Then the experimentally scaled theoretical X-ray production cross-sections have also been acquired by multiplying the Ag L αβγ X-ray production cross-sections in the database of the Monte Carlo code used in this work by the corresponding scaling factor. Some discrepancies between the Distorted Wave Born Approximation (DWBA) predictions and the experimental data provided by different authors have been discussed. In order to analyse the accuracies of the published experimental data with the thick-target method in the literature, the contribution of the multiple scattering of incident positrons, from the bremsstrahlung and annihilation photons and other secondary particles to the characteristic X-ray yields has been estimated by means of the Monte Carlo simulation technique in combination with the theoretical integral calculation.