Search

We present analyses of lead glazes from dishes and storage bowl from six Renaissance (1536 to 1660 CE) sites in southern Denmark and northern Germany. All are high-lead silicates, ranging from 55 wt% to 75 wt% PbO, with only silica and alumina routinely exceeding c 1 wt%. The underlying bodies are ferruginous clays with 15 to 20 wt% Al2O3 and Here, we discuss the relationship between the PbO content in the glaze with the amount of lead entering acidic foodstuffs stored or served in these vessels. The experiments used 4 wt% acetic acid, similar to common foodstuffs preserved with vinegar such as pickles. This concentration is high enough to be effective in preserving food, while low enough to taste good. It also approximates official test methods for the release of Pb for glass hollowware in contact with food according to ISO 7086-1. We found a strong exponential correlation between lead leaching and lead oxide content in the glaze, reaching toxic levels in the leachate solution after short periods of exposure (less than 1 day) for PbO contents above c 70 wt% PbO. Our research shows that widely distributed lead-glazed ceramics had the potential of being a main source of Pb in human bodies. How the pottery was actually used, in combination with kitchen wares fashioned from other materials, might explain why lead-glazed ceramics might not have had quite the poisonous consequences the etching experiments suggest.

Additional file 1. Photos of examples of samples analysed in the present work. Analyses of Corning glass standards for SEM-EDS. Full ternary phase diagram.

International audience; Strength of wine trade during early Roman Empire is evidenced by the wide geographical distribution of Gauloise 4 amphorae crafted in Narbonnensis area. Extension of that trade to Iberian Peninsula has previously been recorded, but remains very poorly understood. Recent discovery in Cartagena of several pottery sherds closely similar to Gaulish productions led us to investigate their provenance. Eight different amphorae sherds were analyzed using X-ray fluorescence and compared to a solid database of more than 700 amphorae samples produced in south of France. Data was then analyzed by a hybrid clustering method combining Principal component analysis and Dynamic cloud algorithms derived from K-means method. Results were subsequently validated with a Factorial discriminant analysis associated to a cross-validation method. The very high probability of attribution we obtained allowed us to conclude that at least five of the eight samples were produced in workshops located in the Narbonnensis area; La vaste distribution des amphores de type Gauloise 4 dans le monde méditerranéen durant le Haut-Empire témoigne de l’intense activité du commerce vinaire de la Narbonnaise durant cette période. La diffusion de ce type d’amphore dans la Péninsule ibérique, attestée par quelques amphores, demeure encore mal connue. La récente découverte à Carthagène de plusieurs fragments de Gauloise 4 présentant de fortes ressemblances avec les productions de Narbonnaise nous a conduits à réaliser une série d’analyses chimiques afin de nous assurer de leur provenance. Ainsi, huit amphores ont été analysées par fluorescence de rayons-X et comparées à la base de données de la Maison de l’Orient de la Méditerranée, riche de plus de 700 résultats d’analyses d’amphores produites dans le sud de la France. Les données ont ensuite été traitées en utilisant une méthode statistique de partitionnement mixte combinant l’Analyse en composante principale et l’algorithme des Nuées dynamiques (dérivé de la méthode K-means). Les résultats ont ensuite été validés par une analyse factorielle discriminante associée à une validation croisée. Les probabilités d’attributions obtenues nous ont permis de conclure qu’au moins cinq des huit échantillons ont été fabriqués dans les ateliers de Narbonnaise