Heinrich Taubald

Universität Tübingen, Institut für Geowissenschaften

taubald@uni-tuebingen.de

In course of former collaboration in the frames of DAAD-MÖB, we could establish excellent working connection between the Hungarian and German partners (Tübingen University, Hungarian National Museum, ELTE University and the Geochemical Research Laboratory of the HAS). The program "Archaeometrical study of Roman and Mediaeval Marbles in Hungary" could solve and clarify the provenance of the marble relics found at Heténypuszta Roman fortress, completed our knowledge on marble quarries in and around the province Pannonia Already during the former joint project, the idea and necessity of joint research devoted to pottery emerged. In continuation to this very successful exchange project in the years 2001 and 2002, we are glad to have again financial support by German DAAD and Hungarian MÖB for a new project to be carried out in 2005 and 2006, this time on pottery and its raw materials in Hungary.

Pottery production is one of the most important crafts of prehistoric communities. Most of the archaeological evidence recovered from excavations consists of sherds and different ceramics, found on habitation sites as well as cemeteries. Pottery therefore has served for long as basic starting point of archaeological analysis of a site. Material characteristics of the ceramics may depend on many complex factors: technological, regional, cultural and functional factors. Our project intends to deal mainly with regional factors comparing local sediments with the material of early ceramics all over Hungary.

Our planned analysis is aimed at systematically clearing of at least one aspect of problems connected with prehistoric pottery making, the regional factor. By this we hope to get nearer to solving the archaeometrical, petrological and historical problems emerging in the study of pottery.

The aim of our work is to compare the mineralogical, petrological and geochemical composition of ceramics and local sediments (clay, silty clay, sand, etc.) which could be raw materials for pottery making, in three different regions within Hungary: Western Transdanubia, South-Eastern Hungary (SE-Alföld) and North-Eastern Hungary. Possible local sources of raw material can be characterised by young river sediments of different water catchment areas (gathering grounds of river Danube and Tisza), as well as piedmont areas characterised by Mesozoic sedimentary and Palaeozoic sedimentary and low grade metamorphic rocks.

Planned research will include investigation of possible raw materials collected on the archaeological sites and from the surrounding areas. Sediments from these geological areas have not yet been characterised geochemically. We are planning to compare the results of analysis of the sediments to the petroarchaeological results obtained from prehistoric pottery.

We will apply state of the art analytical facilities to get the best information possible about the investigated samples:

• NAA (Neutron activation analysis for chemical composition) Hungary
• PGAA measurements (prompt gamma activation analysis for chemical composition) Hungary
• XRD (X-ray diffraction for mineralogical phase analysis) Hungary
• Microprobe if necessary (chemical composition)
• Isotope geochemistry (Sr isotope analysis for identification of source materials) Germany
• XRF (X-ray fluorescent analysis for chemical composition - destructive and non-destructive) Germany

Intensive personal exchange between Budapest and Tübingen will help to get the best out of the analytical measurements.

With the performance of Sr- Isotope measurements in this project we are applying radiogenic isotope techniques to archaeological problems. Sr may help to distinguish between different sherds and raw materials and to correlate them (for further details regarding the application of Sr see PINTER, this volume).

The first, pioneering work, was done in the Geochemical Laboratories in Tübingen by Onno Knacke-Loy in the nineties, who focused on chemical and isotopical studies of Troian sherds. By applying Rb, Sm-Nd and REE techniques he was able to distinguish between different localities of production and proved the import of Mycenaean ware to Troia.

For detailed isotope analysis about 50 mg of sample powder is required (applies for Sr but as well for Nd, or other isotope systems). The samples are digested over night with HF-HClO₄ and then kept at 170°C for another 6 days to dissolve also accessory minerals. Sr is then separated from the solution by adding 5 ml sample solution to a cation-chromatography resin. The separation process of Sr from the other elements is necessary to avoid peak overlapping in the mass spectrometer. All chemical steps are carried out in a ultra-clean lab, in order to avoid contamination of the sample by other Sr-bearing material, since Sr usually shows a relatively low concentration (few ppm) in the samples to be analysed.

Sr isotopes (⁸⁷Sr/⁸⁶Sr ratio) are then measured with a Finnigan MAT 262 Thermion-mass-spectrometer calibrated with international standards. There are four naturally occurring Sr isotopes Sr-88, Sr-87, Sr-86 and Sr-84. Only Sr-87 also has a radiogenic portion, since Sr-87 is also formed by radioactive decay of Rb-87. So, due to different Sr and Rb concentrations and time of accumulation different clay sources probably have different ⁸⁷Sr/⁸⁶Sr ratios. These ratios do not change during transport of raw material, fabrication or firing of sherds and thus can provide a not erasable fingerprint for a clay source.

XRF analysis is a conventional, modern and powerful tool to study major and trace element composition of geological material. It is used in Geosciences for a long period of time and has proved its performance for generations of geologist and geochemists, it is suitable for a couple of interesting elements. Of course not all elements can be analysed with the same precision and accuracy with all methods, but the analytical facilities applied in this project provide the ideal chance to get almost the complete system of elements, since they complement one another in a perfect way.

In former times however, XRF was a destructive method, where 1,5 g sample powder was used to get a fused bead on which the composition was analysed. For more than five years now, there is the possibility to get quantitative chemical information with XRF in a non-destructive way. All you need is a specimen small enough to fit in our sample holder (about 4 cm in diameter) and even enough to cover an 8 mm aperture homogeneously. While in former times only qualitative analysis was possible, now, due to improved analytical performance as well as software skills, quantitative analysis is made feasible by the new facilities.