1.3 Cosmogenic nuclide and Ar/Ar dating techniques and facilities in Scotland

Introduction and explanation of the technique

(Ar =Argon, K= Potassium, Be = Beryllium, Al = Aluminium, Cl = Chlorine, Ca = Calcium, He = Helium, Ne = Neon, Sr = Strontium)

The terrestrial cosmogenic nuclide (TCN) and 40Ar/39Ar chronometric techniques are widely used in Earth and environmental sciences. TCN techniques rely on the ingrowth of nuclides within the mineral lattice (hence, in situ TCNs) as a result of the interactions between secondary cosmic radiation and minerals in that lattice, and the Ar-Ar technique is a development of the technique that relies on the decay of K to Ar to date volcanic rocks and weathering products. Recent technical advances in both fields now allow the techniques to be used on timescales that are relevant to archaeology, and although technically challenging, both techniques are now capable of measuring sub-1,000 year ages. TCNs accumulate at the Earth’s surface and so provide a chronology of exposure (Siame et al. 2006; Dunai 2010) . Recent archaeological applications include the dating of flint artefacts and building structures (Akçar et al. 2008). TCNs can also be used to determine rates of erosion, and multiple nuclides with different half-lives can be used to date the deep burial of materials (e.g., in caves where the materials are cut off from cosmic radiation). Such burial dating is best suited to older settings, however, such as Palaeolithic stone artefacts that have been buried for hundreds of thousands of years, and so is not likely to be useful in the currently understood Scottish context.

A blog of Dr Greg Balco of the Berkeley Geochronology Center in California has a very useful and up-to-date discussion of the issues associated with burial dating and its application (http://cosmognosis.wordpress.com/2010/11/03/exotic-burial-dating-methods/; accessed 15 July 2011). Novel applications of multiple nuclides with different half-lives are also being developed for determining ages of timing and amounts of soil erosion in the past, with potential applications to archaeological settings (see below).

Ar/Ar dating is limited to K-rich minerals, such as sanidine, from volcanic ashes and is primarily used to bracket the timing of site occupation. The SUERC Ar/Ar laboratory has three magnetic sector mass spectrometers which it operates for the NERC-funded national facility.

A survey of the Scottish application and a brief literature review

As far as is currently known, TCNs have not yet been exploited directly in Scottish archaeological contexts, but there is great potential for their application, given the research capacity and analytical capability that are available in Scotland (see below). In Israel, for example, Verri et al. (2005) interpreted the differing 10Be contents of independently dated chert tools in a cave. They argued that the chert for artefacts with very low 10Be concentrations must have been obtained from mines or pits deeper than a few metres (i.e., below the depth of production of 10Be), whereas others artefacts had measurable but variable 10Be contents, indicating that the chert for these artefacts had been obtained from either deposits of varying depths (i.e., varying degrees of exposure to cosmic radiation) or from surface materials that had had varying durations of exposure at the ground surface. In other words, TCN analysis in this context is a valuable tool for provenance studies and other specific applications may yet be developed.

Research capacity within Scotland

There are two types of TCNs, radioactive (‘radio-nuclides’) and stable. The radio-nuclides, primarily 10Be, 26Al and 36Cl, with 14C currently in development[1], are extracted from exposed rocks and minerals, at several dedicated laboratories in Scotland. SUERC hosts the joint Glasgow-SUERC laboratory and, on behalf of NERC, the Cosmogenic Isotope Analysis Facility (CIAF) dedicated to UK scientists. The University of Edinburgh laboratory is in the School of

GeoSciences. 10Be, 26Al and 14C are extracted from quartz, involving two processes: (i) the preparation of ultra-pure quartz from the sample rock, and (ii) the extraction of the relevant nuclide from that quartz, by chemical means in the case of 10Be and 26Al, and by heating in the case of in situ 14C. 10Be and 26Al is extracted and prepared for measurement at all three Scottish facilities, whereas Edinburgh is the only facility in Scotland that has historically had capability for extracting cosmogenic 36Cl (generally from CaCO3-bearing rocks, in which the 36Cl is formed from Ca). SUERC hosts a developmental line for the extraction and graphitization of in situ 14C from quartz, being currently one of probably only three laboratories world-wide that is producing repeatable measurements of in situ 14C in ‘unknown’ samples (Fülöp et al., 2010a). All radio-nuclides are measured using the 5 MV tandem accelerator mass spectrometer (AMS) at SUERC, the only AMS for cosmogenic nuclide measurements in the UK and the only AMS world-wide dedicated solely to geosciences applications.

The stable TCN, 3He and 21Ne, are measured at the SUERC noble gas isotope laboratory. It is equipped with a new HELIX-SFT multiple-collector mass spectrometer capable of highly precise cosmogenic He and Ne isotope determinations.

Emerging opportunities, future research areas, and future needs

The foregoing highlights the fact that Scotland hosts virtually all of the UK’s expertise and preparation facilities for TCN analysis, and certainly all of Scotland’s analytical capability, presenting an unparalleled opportunity for Scottish archaeology.

It has been noted above that there are two principal routine uses of TCN analysis, namely, surface exposure dating and the determination of catchment-averaged rates of erosion. Surface exposure dating is widely used to provide ages of deglaciation for Scotland (e.g. Stroeven et al., 2010) and for the ages of particular landforms (e.g., Fabel et al. (2010) for the famous Parallel Roads of Glen Roy; Stone et al. (1996) for the raised rock platforms of Lismore). There is obvious potential here for dating the emergence of Scottish coastal landscapes from below sea-level for subsequent occupation. Unpublished TCN surface exposure dating of raised beaches on Jura, for example, confirms their Late Pleistocene emergence and the extreme rates of glacio-isostatic surface uplift immediately after deglaciation (Fabel et al., in prep), and many glacial landforms are amenable to the same type of analysis in an archaeological context.

TCNs can also be used to determine catchment-averaged rates of erosion, provided certain conditions can be met (Dunai 2010, 122). Thus, geoarchaeological investigations could in principle include determination of catchment-averaged erosion rates, but it must be remembered that the rates so determined integrate the erosion rate for the time period it takes the material to get from 2m depth in the regolith/soil to the ground surface, where it is detached. New work using nuclide pairs, one of which is in situ 14C, has the potential to quantify the timing and amount of at-a-point Holocene soil erosion (as opposed to the catchment-averaged rates provided by TCNs or, indeed, by sediment flux/mass balance studies). This new work utilises depth profiles of in situ 10Be and 14C in sediments/soils of known age and exploits the different production rates and half-lives of the two nuclides to the timing and amount of Holocene soil erosion (e.g., Fülöp et al. 2009, 2010b). At this stage, the work is laborious, expensive and developmental, but virtually all analytical geochronological techniques are thus as they are developed but ultimately become more affordable as they become more routine.

It is worth recalling here that the application of all geochronological techniques is never fully routine, though some are more widely understood than others and their assumptions (and the associated potential pitfalls) are part of the ‘working toolkit’ of many archaeologists. TCN analysis is still overall in a developmental phase, although it is reasonable to state that its application is becoming more routine. The assumptions and necessary ‘pre-requisites’ for the technique are probably less likely to be routinely well understood by archaeologists. For that reason, it is essential that anyone seeking to employ TCN analysis link up with a practising TCN analyst (or, at least, an experienced user). The wealth of capability and experience in Scotland where by far and away the bulk of UK expertise in TCN analysis resides should make that relatively easy for anyone interested in TCN analysis.

[Note 1] Note that consideration here is of in situ 14C that is formed in the crystal lattice of quartz by spallation of oxygen by secondary cosmic radiation, and not the 14C that is formed in the upper atmosphere by spallation of nitrogen. The 14C formed by spallation of N is oxidised to CO2 that is taken up in photosynthesis and forms the basis of ‘conventional’ radiocarbon dating. That 14C can be thought of as ‘meteoric’ 14C (formed in the atmosphere) and there is likewise meteoric 10Be that was used in the 1980s and 1990s but fell into disuse once the theory and practice of in situ 10Be had been formulated by Lal (1988). The use of meteoric 10Be is currently being revived for landscape evolution studies (Graley et al. 2010; Willenbring and von Blanckenburg 2010).