Chinese celadon shards of the Longquan type, believed to date from the Southern Song dynasty (1127-1279 AD), were excavated in 1934 on Mapungubwe Hill, a 13th century Iron Age site in the Limpopo valley, South Africa. We studied the shards with Raman spectroscopy, x-ray fluorescence (XRF) spectroscopy and x-ray diffraction (XRD). The Raman polymerization index (I p ), calculated from the spectra of the glaze of the shards, indicated a higher firing temperature than expected for the relatively calcium-rich Longquan glazes of the Southern Song dynasty. XRF analysis of the glaze and XRD measurements of the bulk of the shards supported this view and date the shards to possibly the Yuan (1279-1368 AD) or even early Ming (1368-1644 AD) dynasties. These results have an impact on the chronology of the history of the region and therefore call for further research of a comparative nature of other Chinese celadon shards excavated on archaeological sites in Africa, in addition to additional carbon dates of Mapungubwe hill.

No abstract

X-ray Fluorescence Spectroscopy (XRF) has reached the stage where it is classified as a mature analytical technique. The theoretical principles are well understood. In addition modern instrumentation demonstrates enhanced generator and temperature stability. High sensitivity is possible even for light elements and effective matrix correction software is available to the modem day spectroscopist. Apart from its continued applications in researcli and development. XRF has become u routine prcjcess control tool.X-ray Powder diffraction (XRD), on the other hand, has with minor exceptions as in the cement industry, largely remained a research tool, despite being an older analytical technique than XRF XRD has progressed significantly in the past decade from a mainly qualitative technique for the identification of crystalline materials to a quantitative tool with the advance of more powerful software packages. This software has improved instrument control, but also quantification and structure determination using the Rietveld method. Consequently, XRD is rapidly entering the process control environment.In this paper the authors demonstrate, with practical examples from different industrial applications, how combined XRF and XRD use can provide truly quantitative phase analyses. XRF is used to verify XRD data and visa versa.The data obtained in this study clearly illustrate the value that can be added to either technique if XRF and XRD data are used together, and indicate some applications in routine process co.

This paper deals with the thermal behaviour occurring during the formation of lithium borate glasses. Thermogravimetric analysis and differential scanning calorimetry were employed to identify reactions occurring during fusion in pure flux mixtures. The conditions around the identified reactions were reproduced in a muffle furnace to obtain larger quantities of material for further investigation using spectroscopic techniques. It was shown that above 1050 • C lithium borate fluxes volatilize, which could lead to inaccurate analytical results.

During lithium tetraborate fluxing, chromite undergoes incongruent dissolution, in which high proportions of aluminium, magnesium and titanium are dissolved in the flux while some iron remains behind in a residual chromium-rich phase. Increasing the time and flux/sample ratio improve the dissolution of chromite. Errors in XRF analysis due to incomplete dissolution of chromite in lithium tetraborate flux will have the largest effects for chromium and lesser, although still serious, effects for iron.