Author:

Keywords:

Wadi Natrun, al-Barnuj, Lake Pikrolimni, Fezzan, Natron, Glass, Science & Technology, Social Sciences, Life Sciences & Biomedicine, Physical Sciences, Anthropology, Archaeology, Geosciences, Multidisciplinary, Geology, LATE-ANTIQUITY, MARINE, SILICATE, ELEMENTS, AREA, SAND, 0402 Geochemistry, 0403 Geology, 2101 Archaeology, 4301 Archaeology

Abstract:

Boron is a trace element present in natron glass, which largely enters the glass via the flux. Therefore, the B isotope ratio is targeted as a means of provenancing this flux. In this work, 33 Greco-Roman natron glasses and 18 natron samples were analysed for their B isotopic composition (expressed as δ11B). All glasses, except one of mixed flux origin, show a δ11B value close to the average value of+29‰. 11 natron samples from the Wadi Natrun show an average δ11B of+29.8‰, while the sample from al-Barnuj shows a δ11B of+28.5‰, the average for 4 Fezzan samples from Libya is+25.3‰ and that for the 2 samples from Lake Pikrolimni in Greece+10.5‰. There is an influence of the sand on the δ11B of natron glass, complicating the picture, but in general, the silica source will slightly lower the δ11B of the glass compared to that of the natron source. The high melting temperature has no significant effect on the B isotope ratio. It can be concluded that Greco-Roman natron glasses show a rather homogenous δ11B. The Wadi Natrun salts analysed fit a model in which the δ11B of the glasses is somewhat lower than that of the natron source due to the input of low δ11B sand. The samples from Lake Pikrolimni in Greece analysed in this study show too low δ11B values to be used in glass making according to this model. © 2014 Elsevier Ltd.