Many centuries before the rise of nanotechnology, 9th-century Abbasid potters were producing ceramics which had striking colours (golden, red, brown or yellow) and metallic appearance originating from nanoparticles. This technology, which created the iridescence effect, emerged in the Near East during the Abbasid period and subsequently spread across the Islamic world. “Today we still find it very difficult to reproduce the effects they did in the 9th century, so we wanted to find out what chemical transformation the painting applied on the ceramics went through to create such effects”, explains Trinitat Pradell, professor at the UPC and co-corresponding author of the publication.

In order to achieve the effects, artists would start by painting a glazed ceramic surface with a mixture containing metallic compounds, notably silver and copper. During a subsequent firing, these elements diffused from the paint to the glassy glaze and underwent chemical reactions, including reduction, leading to the formation of metallic silver and copper nanoparticles within a thin surface layer. The particles were therefore not simply deposited on the ceramic: they became embedded in the glaze, creating a nanostructured region whose composition and morphology determined how the surface interacts with light.

The chemistry of the firing process was particularly important. The final appearance was not determined simply by the presence of silver or copper in the paint, but by the complex interplay between them and other elements present in the glaze. In particular, the relative amounts of copper and silver influenced the chemical reactions taking place during firing and, consequently, the formation and characteristics of the metallic nanoparticles.

The UPC team joined forces with the ESRF’s Marine Cotte and colleagues to study these ancient materials using X-ray spectroscopy and X-ray fluorescence on ESRF’s ID21 beamline and X-ray diffraction at ALBA synchrotron. . “The micro size of the beam and the study of Fe, Cu and Ag have been essential to unveil the chemistry of the lustre layers, and this would have not been possible without Marine Cotte and the people of ID21 beamline adapting the beamline to our problem” , explains Pradell.  They found that luster is obtained through an ion-exchange process during firing, whereby copper and silver ions diffuse from the paint into the glaze, replacing the alkali ions. To enable this process, the luster must be fired at a temperature between the glass transition temperature and glaze softening, in order to favor ionic diffusivity and prevent the luster paint from sticking to the glaze.

The results also showed that differences in composition and microstructure between different luster colours are primarily due to the relative amounts of copper and silver present in the luster. The concentration of silver and/or copper species and nanoparticles near the glaze surface is controlled by the relative amounts of copper and silver in the paint itself, as well as by the reactions between them. Together, they determine the type, amount, distribution, and size of the nanoparticles.

What comes next? Marine Cotte, scientist at the ESRF and co-corresponding author of the publication, explains: “This study was carried out by analysing tiny fragments from historical ceramics. The next step will be to produce and analyse mock-up samples to distinguish the effects of each manufacturing parameters. We’d also like to explore the role of other metals, such as iron and tin”.

Reference:

Sadurní, R. et al, Science Advances, 9 Sep 2026, Vol 12, Issue 37 DOI: 10.1126/sciadv.aeg55

Text by Montserrat Capellas Espuny