Dr Juliet Spedding, Research Assistant on the Craft Interactions Project in the Department of Scientific Research at the British Museum and an Honorary Research Associate within the Department of Archaeology, Classics and Egyptology, has continued her research utilising a novel method to analyse ancient Egyptian faience without damaging invaluable artefacts.
Following on from her 2024 analysis using a particle accelerator to analyse ancient objects, Juliet returned to the ISIS facility in Didcot, Oxfordshire in 2025 for two further experiments to examine faience – the first man-made non-clay ceramic that consists of a quartz/silica pale coloured core with a bright shiny glaze over the top which is usually a turquoise blue. For this visit, however, Dr Spedding examined the quantities of raw materials used and carried out analyses on objects that had not previously been investigated by other methods.
The ISIS Neutron and Muon Source, part of the Science and Technology Facilities Council (STFC), granted her five days beamtime for two separate visits, one in May and one in October, which allowed her to analyse five objects during each visit. The longer timeframe was necessary as each object took a whole day to examine, with three analyses at different depths required. As with previous experiments, each object was placed in or wrapped with aluminium foil before being placed in the beam. Dr Spedding was also able to work with her collaborators from the ISIS facility, Dr Adrian Hillier and Dr Sayani Biswas.
During the May analysis, the five objects analysed raised some interesting questions about objects found at the same site and how the different raw materials used may have been influenced by the finished object type. It showed the potential for the use of both a sandstone-based raw material and a purer raw material for the core of different faience objects found at the ancient royal capital of Meroe (Sudan). For example, a sistrum handle using a more crumbly sandstone-based core versus a piece of architectural faience that had a bright white, very pure core that did not crumble . The most exciting object to investigate, however, was a complete royal shabti from Nubia (ancient Sudan), which had never been chemically analysed before.
Photo: Shabti of Anlamani from Nuri (courtesy of the Garstang Museum, University of Liverpool)
This is a complete faience shabti of the ancient Sudanese ruler Anlamani (c.620-600BCE) from his pyramid at the site of Nuri. A shabti was a funerary object designed to perform tasks for the deceased in the afterlife. The first analysis examined the core of this object and showed it to be very pure silicon with only iron as the impurity. It also did not show any copper which—as this shabti is a turquoise blue—is the most likely colourant. The use of a purer silicon source for the core would fit the royal status of the owner of this object and whilst it might be expected that the best materials would be used sometimes analysis of these objects shows something else was used. The second analysis, performed closer to the surface, again indicated the presence of silicon and iron but no copper. Finally, an analysis was completed overnight at just below the surface which showed the presence of silicon and iron and potentially copper. This allowed an initial identification of the glazing technique used for this object—cementation. This method involves the burying of the object(s) to be glazed in a glazing powder which adheres to the surface when the object is fired.
To obtain these detailed analytical results you would usually need to destructively sample the object, something not possible with an artefact like this shabti. However, by using muons and their depth-profiling capability, it is possible to analyse ancient objects non-destructively.
When Juliet returned in October 2025 she examined a further five objects. This presented the opportunity to compare objects from similar sites to investigate potential similarities and/or differences. The results raised the possibility of a more sandstone-based core used in a faience lion from Meroe that depicted the Nubian lion-headed god Apedemak, again showing the potential different sources for the core of these ancient objects at the site of Meroe. This object is incomplete but a similar depiction of the god Apedemak as a lion can be found in the Garstang Museum.
As a result, these findings highlighted the many different potential core compositions that were being used for faience found at a single site. Similarly, preliminary results of two pieces of faience from the turquoise and copper-mining site of Serabit el-Khadim, Sinai Peninsula, one in May and the other in October, allowed a comparison of material from the same site. The results show the potential for different raw material sources, which may be due to different production centres for these objects, indicating that there may have been a variety of sources for the faience found at this site outside the Nile Valley.
Photo: Fragment of a statue of a hippopotamus (courtesy of the Garstang Museum, University of Liverpool)
Dr Spedding also analysed part of an incomplete hippo statuette (E.128), which are well known objects from the Middle Kingdom (c.2050–1650BCE), the most famous example being ‘William’ in the Metropolitan Museum of Art, New York. The results of the analysis of this object showed that the material used for the core was not pure silica due to the presence of iron, despite the white appearance of the core which may have indicated otherwise.
The use of muons for faience analysis is still in its early stages, but through further examination of objects using this method it is hoped that this will enable the creation of a database of results of these delicate artefacts - allowing us to answer some questions about the production of this fascinating ancient material.
Thanks to the ISIS Neutron and Muon Source, part of the Science and Technology Facilities Council (STFC), for approving this experiment. Also many thanks to Dr Adrian Hillier and Dr Sayani Biswas for all their help before, during, and after the experiment. Thanks also to the Garstang Museum for allowing access to the objects analysed.