Faculty of Physics and Astronomy University of Heidelberg Diploma thesis in Physics submitted by Bente Philippsen born in Heide, Germany 2008 Hard Water or High Ages? 14C food crust analysis on Mesolithic pottery from Northern Germany This diploma thesis has been carried out by Bente Philippsen at the Institute for Environmental Physics under the supervision of Prof. Werner Aeschbach-Hertig. Abstract The aim of this thesis is the investigation of the hardwater effect and its implications for the dating of the first occurrence of pottery in Northern Germany. The hardwater effect denotes the effect of a high carbonate content of freshwater on radiocarbon dating: the dissolved minerals can lead to spurious, too high ages of samples from freshwater systems. The possibility of the hardwater effect in food crusts on pottery is investigated for two sites of the Late Mesolithic Ertebølle culture. The first radiocarbon dates from the food crusts from these sites were surprisingly high. I will examine if the pottery really is so old or if the hardwater effect is responsible for this. The samples I thus have to analyze are very small. Instrumental development was needed for the dating of these samples. On-line combustion combined with stable carbon and nitrogen isotope measurements is developed to minimize the number of steps in the preparation and measurement of a sample so a reduction of contamination may be achieved. Also the conversion of the CO2 from combusted samples to graphite, the form of carbon in which the radiocarbon content can be measured, is examined and some improvements are suggested and tested. Abstract Das Ziel dieser Arbeit ist die Untersuchung des Hartwassereffekts und seines m¨oglichen Einflusses auf die Datierung der fr¨uhesten Keramik in Norddeutschland. Der Hartwassereffekt bezeichnet die Auswirkungen eines hohen Karbonatgehalts von Fl¨ussen und Seen auf die Radiokarbondatierung: die gel¨osten Minerale k¨onnen zu falschen, zu hohen Altern von Proben aus S¨ußwassersystemen f¨uhren. Die Wahrscheinlichkeit des Hartwassereffekts in Speiseresten auf Keramik, in der S¨ußwassernahrung zubereitet wurde, wird f¨ur zwei norddeutsche Fundstellen der sp¨atmesolithischen Ertebøllekultur untersucht. Die ersten Radiokarbon- datierungen von Speiseresten auf Keramik dieser Fundorte ergaben erstaunlich hohe Alter. Ich werde unter- suchen ob die Keramik wirklich so alt ist oder ob der Hartwassereffekt daf¨ur verantwortlich ist. Die Proben, die ich daf¨ur analysieren muss sind sehr klein. Um diese sehr kleinen Proben datieren zu k¨onnen, ist instru- mentelle Weiterentwicklung n¨otig. Zum Beispiel soll Fraktionierung minimiert werden. Die Verbindung der Probenverbrennung mit der Messung von stabilen Isotopen wurde entwickelt um die Anzahl der Probenauf- bereitungsschritte zu minimieren. Dieses kann zu einer Reduzierung von Verunreinigungen der Probe f¨uhren. Die Umwandlung des CO2 zu Graphit, der Form von Kohlenstoff in der der Radiokarbongehalt gemessen werden kann, wird untersucht. Einige Verbesserungen dieses Prozesses werden vorgeschlagen und getestet. Preface The influence of the hardwater effect on pottery dating, the necessary methodological and instrumental developments, and the influence the hardwater effect has on the dating of the first pottery in Northern Germany will be presented in this thesis. The hardwater effect is the influence of dissolved carbonate on samples from freshwater systems. It is a long-known and broadly accepted phenomenon. However, the possibility of a hardwater effect in one sample type is still under debate. Food crusts on pottery in which freshwater food such as fish was cooked could as well show spurious ages if the freshwater food came from water systems with a high carbonate content. So far, this effect has not been agreed on. I have therefore chosen two sites next to freshwater rivers in Northern Germany where a hardwater effect on food crusts on pottery is probable. I have taken a large number of archaeological samples as well as examined recent freshwater samples in order to obtain statistically significant results. As the food crust and fishbone samples from these sites are very small, development in sample processing methods was necessary. A device combining sample combustion for AMS 14C dating with on-line stable isotope measurements was advanced and tested. This device has the potential of reducing the number of steps in sample preparation, the contamination risk, and the total sample size. Due to various problems with the ion sources and tandem accelerator, some of the samples could not be measured before this work had to be finished. A new ion source was constructed and it was planned that my samples could be measured using this new ion source. Because of several delayals, it was not possible to start the new ion source for standard 14C datings before this thesis had to be finished. Therefore, the old ion source was used for my samples. There are two disadvantages with that: First, the old ion source is not able to produce enough ion beam current for dating when the samples are very small. Second, the ion source is only optimised for samples graphitised with cobalt at 700◦C. My small archaeological samples could thus not be measured at all. Normal-sized samples graphitised with iron at 550◦C are measured with a far higher uncertainty than would be the case in the new ion source. Working in the intersection of physics and archaeology The topic lies in the intersection of physics and archaeology, an area which is also called archaeometry. • “Archaeology is the study of the human past through material remains” (Hayashida 2003). • “Physics is the study of matter and energy and how they work with each other” (http://simple.wikipedia.org/wiki/Physics on February 26, 2008). Archaeometry is thus the (e.g. physical or chemical) analysis of material remains to study the human past. Archaeometry is another term for archaeological science and comprises different scientific technologies which are applied in archaeology. This can be chemical and physical dating methods, provenance studies of arte- facts, study of the distribution and use of artefacts, reconstruction of past landscapes, climate and envi- ronment, dietary studies of humans and animals, remote sensing and geophysical surveys, and conservation techniques (Tite 2003a). This work will deal with two aspects of science applied to archaeology: radiocarbon measurements for dating and stable isotope measurements on food crusts for “dietary” studies. Instumental development I and examination of the existing sample preparation methods will form the physical part of the work. At the same time, it will also deal with the archaeological background on a scientific basis, bearing in mind the appeal of Dunnell (1993): “The attraction, particularly from the physical sciences side, often seems to be to the non-scientific archaeology of the Sunday newspapers’s feature section”, and “there will be no basis for integrating archaeometry and archaeology until archaeometrists focus their attention on scientific archaeology”. Tite (2003a) also emphasizes this fact: “To ensure that archaeometry remains relevant to archaeology, it is essential that only real archaeological questions are addressed. This in turn necessitates the maintenance of a substantial dialogue between archaeometrists and archaeologists together with a holistic approach that goes beyond reconstruction to a full interpretation within the specific archaeological context under investigation.” On the other hand, archaeologists should be careful not to depend too much on scientific dates. It is tempting to use for example radiocarbon dates as a calendar for prehistory, thus giving archaeologists the role of historians. But it has to be kept in mind that reporting the archaeological phase of a site is often more useful for comparing findings in a larger region than expressing the time horizon in centuries BC. Absolute chronology may be the aim of archaeology, but it must never be the backbone in a way that a calendar is the backbone of history (Fischer 1976). The topic presented in this thesis was especially interesting to me as I had gained insight both in physics and prehistoric archaeology in my studies. During archaeological studies and in physics projects, I could learn about physical dating methods and especially radiocarbon dating. In archaeology, I specialized in the Neolithic in Northen Germany and South Scandinavia, particularly in the Mesolithic-Neolithic transition. This made it easier for me to take both the physicist’s and the archaeologist’s point of view when discussing the sites I examined and their importance in the cultural development of the region. I hope therefore that I can avoid some of the possible mistakes that happen when two differnent disciplines meet. Outline of thesis The first chapter presents an overview over the methods applied. After an introduction to the basic principles of AMS radiocarbon dating and stable isotope analyses I will describe the method of pottery dating. In the second chapter, I will present the development of enhanced techniques for the preparation of small samples. The third chapter will deal with the sites from where I got the samples as well as their geological and archaeological background. In the same chapter, the events being dated in this thesis will be put into the context of cultural development at that time. The fourth chapter will finally deal with the archaeological samples and the recent test samples which all were used to determine the hardwater effect for the two rivers. Many of the physical terms may be unknown to archaeologists and, the other way round, physicist are likely to be unfamiliar with the archaeological terms. A glossary at the end of this thesis may contribute to a better understanding for specialists of both fields without lengthening the text. Acknowledgements First of all I would like to thank the AMS 14C Dating Center of the University of Aarhus and its director, Jan Heinemeier, for a warm welcome and the opportunity to work on an exciting topic. I was in the fortunate position to benefit both from the laboratory’s equipment and its know-how as well as from its contacts to scientists of other disciplines.