Cooking pot handle fragment V.008.2/2948.1.
Christian. Degrigny (HE-Arc CR, Neuchâtel, Neuchâtel, Switzerland) & Lucile. Ruynat (HE-Arc CR, None) & Valentin. Boissonnas (HE-Arc CR, Neuchâtel, Neuchâtel, Switzerland)
The object is the tip of a cooking pot handle representative of the Augustan age (100 AD), recognisable by the axial symmetry and the two orifices surrounded by prominent edging and flared shape. The upper part is composed of an openwork flower with three petals, comparable to the representation of the fleur-de-lis. A curved pattern frieze can be seen in the centre of the object. No decoration was found on the back. Before treatment, the object was covered with green corrosion products and organic remains, most of which were mineralised (Figs. 1a and b). Dimensions : L = 73 mm ; W = 48 ; T = 9 mm ; WT = 80.8 g.
Household implement
Romans legionnaires' camp of Vindonissa (present-day town of Windisch), Aargau canton, CH., Windisch, Aargau, Switzerland
2008
Roman Times
1st century AD
Soil
Archaeological Service of the Aargau canton, Brugg, Aargau
Archaeological Service of the Aargau canton, Brugg, Aargau
V.008.2/2948.1. (Inv. Number at HE-Arc: 2001)
2008: Probably a first sediment clearing by archaeologists during the excavation. 2016-17: Lucile Ruynat, mechanical removal of the corrosion products over the limitos with ultrasonic scalpel. Conservation of wooden remains at the top right of the front of the object. Protective layer (varnish) with acrylic resin Paraloid B72®.
The soil of Windish is on the border between calcareous soil (Jura) and molasses-type (Swiss plateau). Calcareous soils tend to be alkaline whereas the molassic soil is generally more neutral. The Swiss climate is temperate, with the four marked seasons. It can have strong frosts in winter.
The stratigraphy below gives an overview of the corrosion layers encountered on the object from visual macroscopic observation. The stratigraphy was created before treatment by observation under a binocular microscope and modified during the mechanical removal of corrosion products.
No sample from the metal was possible. Only a few invasive samples were taken for analysis of corrosion products, as indicated on figure 2 (red squares 1 and 2).
Leaded Bronze
As-cast
None
None
Archaeological Service of the Aargau canton, Brugg, Aargau
2017, chemical analyses
None.
Analyses performed:
XRF with portable X-ray fluorescence spectrometer (NITON XL3t 950 Air GOLDD+ analyser, Thermo Fischer®). XRD of powder samples using Stoe Mark II-Imaging Plate Diffractometer System (Stoe & Cie, 2015) equipped with a graphite-monochromator. Mo-K𝛼 radiation (𝛾 = 0.71073Å, beam diameter 0.5 mm, exposure time: 10 min). SEM/EDS on the object. Raman on powder samples of corrosion products. SEM on sample of organic remains, as well as X-ray picture not presented in this report.
XRF analysis carried out after the cleaning process (Fig. 2, blue dot 3) showed that the metal is a leaded bronze (Table 1). Theoretically the metal is as-cast and should present a dendritic structure.
Elements mass % |
Cu |
Sn |
Pb |
Sb |
Si |
P |
Ti |
Fe |
M1 |
61.1 |
21.1 |
14.2 |
0.9 |
0.9 |
0.4 |
0.2 |
0.1 |
Table 1: Chemical composition of the metal. Method of analysis: XRF, mode General metals, 60s (filters M20/Lo20/Li20). Located at point 3 Figure 2, UR-Arc CR.
XRF analyses of CP4 (Fig. 2, blue dot 5) and CP5 (Fig. 2, blue dot 4) are given in Table 2.
Elements mass % |
Cu |
Sn |
Pb |
Sb |
As |
Ag |
Fe |
P |
Cl |
S |
BAL |
Dark-red layer (CP4) |
24.9 |
18.2 |
11.2 |
0.7 |
0.1 |
0.09 |
0.1 |
0.3 |
0.4 |
7.2 |
36.1 |
Red area layer (CP5) |
33.5 |
14.5 |
16 |
0.5 |
0.8 |
0.1 |
0.05 |
1.3 |
0.6 |
1.5 |
30.5 |
Table 2: Chemical compositions of the dark-red and the red layers. Method of analysis: XRF, mode mining Cu/Zn, 180s (filters M30/Lo30/H60/Li60). BAL corresponds to the elements not analysed: O and C, UR-Arc CR.
The mechanical removal of corrosion products stopped at the limitos, so the metal was not directly observed. Based on Table 1, it seems to be a leaded bronze.
Dendritic structure
Cu
Sn, Pb
None.
Above the well-preserved metal core, CP4 is discontinuously interlocked in CP5. The XRD analysis done on a sample (Fig. 2, red square 2) of both corrosion products, shows that they are mainly constituted of cuprite (Cu2O), (Fig. 5). The colour difference can be explained by a tin enrichment in the darker CP4, as measured by XRF (Table 2). The EDS analysis of the light green CP3 (red square 1 Fig. 2) layer did not reveal the presence of chlorides (Fig. 6). Complementary XRD analysis validated the absence of nantokite. Indeed the XRD and Raman indexing was not successful probably because the compound has a large amorphous part. The other layers were not analysed, we expect CP1 and CP2 to be malachite (copper carbonate hydroxyde) because of the green colour and absence of chlorides. Organic remains were mineralised and preserved by the corrosion process.
Uniform
Mostly type I with locally type II (Robbiola)
None.
The schematic representation of corrosion layers of Fig.3 integrating additional information based on the analyses carried out is given in Fig. 7.
The object is a leaded bronze with a well-preserved metal core. Covering the metal are two strata (CP4 and CP5) composed of cuprite, the darker (CP4) appears enriched in tin. The superior interface of these layers represents the limitos. In areas the limitos, CP4 and CP5 have been replaced by a light green porous corrosion product (CP3). The following layers CP2 and CP1 are probably malachite.
The powdery green corrosion layer CP3 has frequently been observed on bronzes of Vindonissa where it can be located within cuprite or malachite layers. It is typically developed below the limitos and renders the latter extremely fragile. The nature of this corrosion product has not yet been determined. However, the absence of chlorine indicates that it is not a chlorinated corrosion product such as atacamite or paratacamite. The aggressive urban soil could be a reason for the transformation of these naturally grown and stable corrosion layers.
References on object and sample
1. Käch. D. (2012) Jahresbericht 2011, Gesellschaft Pro Vindonissa, Brugg.
2. Käch, D. (2013) Jahresbericht 2012, Gesellschaft Pro Vindonissa, Brugg.
3. Holliger, C. (1984) “Bronzegefässe aus Vindonissa », In: Jahresbericht, Zürich, p.47-70.
4. Martin, M. (1994) Objets quotidiens de l’époque Romaine, Musée Romain d’Augst, Augst.
References on analytic methods and interpretation
5. Feugere, M. (1994) « La vaisselle gallo-romaine en bronze de Vertault (Côte-d'Or) », In : Revue Archéologique de l'Est et du Centre-Est, p.137-166. [Consulté le 22.03.2016] https://halshs.archives-ouvertes.fr/halshs-00580295/document
6. Montandon, B. (1997) Le travail du bronze à l’époque Gallo-Romaine, In Chronozones, n°3, p.2-11.
7. Picon, M. et al. (1996) « Recherches techniques sur des bronzes de Gaule romaine ». In: Gallia, tome 24, n°1, p. 189-215.
8. Robbiola, L. et al. (1998) «Morphology and mechanisms of formation of natural patinas on archaeological Cu-Sn alloys», In Corrosion Science, N°40, p. 2083-2111.
9. Robbiola, L. (1990) « Caractérisation de l’altération de bronzes archéologiques enfouis à partir d’un corpus d’objets de l’Age du bronze. Mécanismes de corrosion ». Thèse de Doctorat, Université Pierre et Marie Curie - Paris VI, Paris. In HAL archives ouvertes [En ligne]. HAL, 2010 [Consulté le 03.06.2017] https://tel.archives-ouvertes.fr/tel-00495356/document
10. Scott, D.A. (2002) Copper and bronze in Art, corrosion, colorants, conservation. Getty publications, Los Angeles.
11. Scott, D.A. (1987) « Metallography of Ancient Metallic Artifacts ». Institute of Archaeology, Summer Schools Press, London, pp. 3-17 and 128-132.
12. Selwyn, L. (2004) Métaux et corrosion, un manuel pour le professionnel de la conservation. ICC-CCI, Ottawa.