Thèse de doctorat
Résumé : Glass is an extraordinary material that has been used for millennia, combining refinement with practical properties. Through the ages, glass production developed and improved considerably. During the Roman Empire, glassmakers were highly skilled craftsmen: not only did they produce delicate vessels, but they were also able to control the production parameters and achieve a large range of colours. How such colours were obtained has unfortunately been lost and we are left without sufficient archaeological and written evidence to understand Roman glass technology. Recently, archaeological glass research has been a hot topic mainly because analytical methods are becoming increasingly available and portable; at the same time, new concepts and ideas are developed. Yet, there is a lack of practical understanding of ancient techniques. Comparisons between archaeological glass science and modern glass chemistry have been limited and conclusions from different fields can even be contradictory. A central objective of this doctoral work is thus to bridge a gap in our understanding of modern and ancient glass.In particular, the focus of this research is set on glass colour. The natural presence of iron as an impurity causes glass to appear bluish. Through the addition of manganese and/or antimony, it is possible to get rid of this hue. However, manganese can either decolour glass or impart a purple colour. It is striking that Roman glass containing manganese in similar concentrations can appear either colourless or purple. How iron, manganese and antimony were mastered by Roman glassmakers to colour and decolour glass is therefore the main research question of this doctoral work. Interactions between these elements are referred to as reduction-oxidation reactions (redox). Their study is central in glass chemistry nowadays.In this doctoral work, the redox interactions between iron, manganese and antimony are thus analysed in-depth using mainly in-situ X-ray Absorption Near Edge Spectroscopy (XANES) and optical spectroscopy. Ancient glass was analysed along with glass produced in a modern laboratory, where various production parameters were systematically controlled. Glass colours were directly related to iron redox and chemical composition. This allowed us not only to observe important differences between ancient and modern glass production, but also to formulate hypotheses about how ancient glass would have been coloured or decoloured. Finally, an assessment of how glass containing iron, manganese and/or antimony is integrated within an archaeological context is addressed. For this, glass from the site of Oudenburg (Belgium) has been analysed.