corundum, ruby, Burma.12, Thailand.12, origin determination, chemistry.2, gems

Examines the potential of proton-induced X-ray emission (PIXE) analysis to distinguish Burmese from Thai rubies through their trace-element compositions, supplementing identification by color and inclusions. Sixty specimens were studied: 29 purchased as Burmese rubies in Thai wholesale markets and 31 obtained from cutters in Chanthaburi and initially regarded as Thai. Microscopic examination supported the proposed origins of 57 specimens, while three lacked diagnostic inclusions. Measurements employed 2 MeV protons from a Van de Graaff accelerator at the National University of Singapore, a silicon-lithium detector and a counting time of 150 seconds per specimen. Samples required only cleaning and mounting. X-ray spectra were processed using the AXIL least-squares fitting program, and calibration standards prepared in an aluminum oxide matrix allowed concentrations to be expressed as weight percentages relative to Al₂O₃. More than ten trace elements were detected, with chromium and iron producing prominent signals; other constituents included silicon, sulfur, chlorine, potassium, calcium, titanium, vanadium, manganese and gallium. Vanadium and iron provided the clearest distinction between the two groups. All initially identified Burmese specimens displayed detectable vanadium peaks, whereas all but one presumed Thai specimen contained very little vanadium. The exceptional specimen also had the lowest iron concentration among the presumed Thai stones and lacked diagnostic inclusions. Its Burmese-like composition led the authors to reassign it provisionally to the Burmese group, producing final groups of 30 specimens each. Average vanadium concentrations were approximately 0.048% in the Burmese group and 0.002% in the Thai group, while average iron concentrations were approximately 0.122% and 0.444%, respectively. Burmese rubies also had higher average concentrations of the other tabulated impurities, although substantial variation occurred within each group. A plot combining vanadium and iron concentrations separated the studied populations more effectively than either visual color assessment or iron content alone. The authors present these compositional relationships as preliminary criteria, emphasizing that additional specimens are needed to establish reliable boundaries and confidence levels. PIXE offers simultaneous multielement measurement, sensitivity at parts-per-million levels for many elements, and analysis without conductive coating, stringent surface preparation or induced radioactivity. Its principal limitation is shallow sampling: the technique measures surface layers rather than the stone’s interior, and near-surface inclusions may distort concentrations interpreted as representative of the ruby itself.