Jingjun Zhu

In 2019, a joint team from Sichuan University, China, and the University of Kelaniya, Sri Lanka, excavated the Mantai port site, unearthing numerous pottery sherds. Traditional typological analysis suggests that these sherds are associated with diverse cultural contexts. However, typological analysis, which relies mainly on visually observable features, generally provides only broad, macroscopic classifications. Moreover, most excavated sherds are fragmentary and lack diagnostic features, which may lead to misclassification. To address this issue, 44 representative pottery sherds were analyzed by proton-induced X-ray emission (PIXE) combined with elastic backscattering spectrometry (EBS) for elemental analysis. Factor analysis of the elemental data revealed two compositionally distinct and internally consistent clusters within the complex assemblage, corresponding respectively to Roman Red Ware and Roman Black Ware, and identified several possible typological misclassifications among visually similar Roman and Indian wares. The comparatively homogeneous elemental compositions of these two groups suggest the use of relatively consistent raw material recipes. Microstructural observations and X-ray diffraction (XRD) analyses also identified two sherds with mineralogical characteristics distinct from other samples. This multi-technique approach refines classification of visually ambiguous pottery and provides a useful analytical framework for interpreting pottery from culturally complex trade-port contexts.

There are two methods for implementing the β-decay Induced X-ray Spectroscopy (BIXS) analysis for tritium contents and depth profile in materials. One is the analytical BIXS method, which can provide tritium content information quickly, but cannot accurately consider the multiple scattering of electrons and geometric complexity. The other is the BIXS method based on Monte Carlo (MC) simulations, where the X-ray energy spectra generated at different depths need to be firstly obtained through the MC simulations, but the MC simulations are very time-consuming. In this paper, a semi-analytical model is proposed to greatly reduce the calculation time of MC-based BIXS method and simultaneously retain the calculation accuracy. In this model, the energy and angle distributions of electrons inside the tritium-containing sample at different depths are calculated by MC simulations, and then combined with the differential cross-sections and shape functions of bremsstrahlung radiation to calculate the corresponding external bremsstrahlung (EB) energy spectrum. The total semi-analytical X-ray energy spectra are compared with that obtained by MC simulations and found that they are in excellent agreement in both amplitude and shape, within 1% difference, and the calculation time is greatly reduced compared with the MC simulations. Furthermore, the semi-analytical BIXS method proposed in this paper are applied to the previous experimental data and observed that the tritium content and depth profile obtained using the semi-analytical BIXS method are consistent with the results obtained by MC-based BIXS analysis, within 0.82% difference.

Absolute measurements were conducted to study the bremsstrahlung emission from ~5-25 keV electrons incident on three thick solid conductive compounds of MoSi$_2$, TiB$_2$ and ZrB$_2$. The additivity approximation was applied in the Monte Carlo PENELOPE simulations for compounds and mixtures. The results showed that in general the experimental bremsstrahlung spectra were in good agreement with the Monte Carlo simulation results, suggesting the feasibility of the additivity approximation in Monte Carlo simulations for the studied cases even in the absolute measurements and that the significant differences between experiments and Monte Carlo simulations near the Duane-Hunt limit for insulating targets in previous studies do not appear in the present studies.

An // neutral particle analyzer (NPA) has been designed and is under development at Sichuan University and Southwestern Institute of Physics. The main purpose of the // NPA is to measure the distribution function of fast ions in the HL-2A/3 tokamak. The // NPA contains three main units, i.e. the stripping unit, the analyzing unit and the detection unit. A gas stripping chamber was adopted as the stripping unit. The results of the simulations and beam tests for the stripping chamber are presented. Parallel electric and magnetic fields provided by a NdFeB permanent magnet and two parallel electric plates were designed and constructed for the analyzing unit. The calibration of the magnetic and electric fields was performed using a 50 kV electron cyclotron resonance ion source (ECRIS) platform. The detection unit consists of 32 lutetium-yttrium oxyorthosilicate (LYSO) detector modules arranged in two rows. The response functions of , hydrogen ions ( , and ) and for a detector module were measured with Am, Cs and Eu sources together with the 50 kV ECRIS platform. The overall results indicate that the designed // NPA device is capable of measuring the intensity of neutral hydrogen and deuteron atoms with energy higher than 20 keV.