Change in silica sources in Roman and post-Roman glass

Although Roman and post-Empire glasses found in Europe are reputed to have a very constant composition and hence source of components, it appears that some 4–5th century and later specimens show evidence of a different source of silica (sand) component. Zirconium and titanium are the discriminating elements. Data presented here for 278 specimens from 1st to 4th century German and Belgian samples indicate a strongly homogeneous Zr and Ti content; N: number of analyzed samples while 62 samples from Maastricht show low Zr–Ti contents from 1st to 3rd century samples while 4–5th century samples show a strong trend of concomitant Ti and Zr increase. If the high values of Zr–Ti represent a new source of silica (sand) the trend from low to high content suggests that a significant amount of low Zr–Ti glass was recycled to form these glass objects. Similar high Ti content can be seen in analysis results reported for other but not all 4–5th century samples found in northern Europe while earlier productions show typical low Ti contents. Although the fusing agent for these glasses seems to have always been natron (a mineral deposit in the Nile delta) from Hellenistic times to the 9th century, a change in the silica source, indicated by variation of the Ti and Zr content, could very well reflect the results of political instability of the 4–5th century exemplified by the fragmentation of the Roman Empire into two parts.