Caas
The Production of Bronze Geistingen Axes
Janneke Nienhuis (Materials Science & Engineering, TU Delft)
Special Axes
Late Bronze Age Geistingen axes are socketed bronze axes of which the function is still a mystery for archaeologists.The axes are socketed, but show no traces of being hafted. Their cutting edges are sharp, but show no use traces. They look like axes, but their thin walls prevent the tools from being used as such. They seem to be part of a serial production, but all have a different chemical composition. And last, but not least: all are deposited in the ground, seemingly without ever being used as tool or weapon. To get a better understanding of these axes, this masters’ thesis in ‘Materials Science and Engineering’ has the research question: “How are Geistingen axes made?”
Fig. 1: Pictures of the Geistingen axes. Top: Nijmegen (AC20, length ± 12.5 cm.); bottom: Tongeren (BH76, length ± 12.5 cm.).
Composition
Samples from two Geistingen axes (fig. 1) are available for research. These are examined with optical and electron microscopes, X-Ray Fluorescence and X-Ray Diffraction.
The axes have a different composition: the Nijmegen axe is identified as a binary copper-tin bronze, while the Tongeren axe is regarded as a ternary copper-antimony-nickel alloy (see table 1). Both axes possess a porous dendritic microstructure with an interdendritic phase, which is typical for a cast material (see fig. 2). In between these phases, Cu2S- and lead-antimony particles are present. The axe from Nijmegen also contains silver particles in addition to the previously mentioned ones. It is assumed that these particles all originate from the raw material, namely the ores used, and that they are formed in the melt.
Fig. 2: Microstructures of the Geistingen Axes. Top: Nijmegen; bottom: Tongeren. Electron micrographs.
Raw material
It is suggested that the same type of raw material is used for both axes, but in different quantities. Suppose there are three ore-types available (see table 1): fahlerz, cassiterite and cuprite. When these ores are mixed with a fahlerz:cassiterite:cuprite ratio of 3:1:3, the composition of the Nijmegen axe can be obtained. Mixing the raw material with a ratio of 50:1:9 results in the composition of the Tongeren axe.
Table 1: Hypothetical composition in at% of three types of ore and the resulting axe compositions in at% after mixing these ores with two different ratios.
Cooling and working
The melt of both axes has been at least at a temperature of 1150 °C, based on the presence of Cu2S-particles. Probably, both objects are water-cooled after pouring the liquid in the mould, according to the secondary dendrite arm spacing. This spacing indicates that the two axes are formed with different cooling rates. No subsequent working of the axes has taken place on the sampling locations.
In conclusion, the two Geistingen axes are produced by co-smelting the same types of ore in different ratios. The bronze is consequently cast into a bi-valve mould. This is then water-cooled to let the object solidify. No subsequent hot- or cold-working has taken place.
The whole thesis can be found in the Online Repository of the TU Delft: repository.tudelft.nl/view/ir/uuid%3A68dd2bd5-97ef-42ab-be75-f13da4bb5f11/
Further Research
Results of further research show that casting of both axes probably took place in a bronze mould. The axe from Nijmegen is identified as a ternary copper-10%tin-3%nickel alloy, while the axe from Tongeren is regarded as a ternary copper-4%antimony-6%nickel alloy. Even though it is still assumed that both axes are made from different raw materials, it is still unclear whether this should be attributed to smelting or re-melting scrap.
More details can be found in 'The Production of Bronze Geistingen Axes' (Nienhuis, in preparation).




