The production rate of 26Al (T½ = 720,000 years) in quartz at the earth's surface is a factor of 6 higher than 10Be. Nevertheless, 10Be is much more widely used in exposure-dating applications. Two factors contribute to this preference. First, the 26Al/Al ratio is intrinsic to the purified quartz, and in contrast to the 10Be/Be ratio, cannot be manipulated by varying the sample mass or the mass of stable isotope carrier. Secondly, beam currents of 27Al- ions from Al2O3 samples in AMS ion sources are generally in the range 100-200 nA, although currents as high as 1-2 μA have been reported. By comparison, 9BeO- currents of several μA are readily achieved. These low 27Al- beam currents in particular have inhibited the wider application of 26Al in exposure-dating applications despite its obvious attractions.
The atomic Al- negative ion has been used almost universally for AMS measurement of 26Al because this choice eliminates any interference from 26Mg. Magnesium does not form a stable negative ion. Beam currents of the AlO- molecular ion are, however, a factor of typically 20 higher than for Al- ions, but have not been exploited because magnesium does form the MgO- molecular ion, and hence the rate of 26Mg ions at the ionization detector completely overwhelms the detector.
The Munich group have already shown that it is possible to use the AlO- ion provided that a gas-filled magnet (GFM) is used to suppress the large flux of 26Mg ions (Arazi et al., NIM B223-224 (2004) 259). We are following the same approach, using an ex-nuclear physics Enge split-pole spectrometer as a gas-filled magnet instrumented with a new multi-element gas-ionization detector. A reduction in counting rate of 26Mg ions by a factor of 1000 is obtained from the GFM alone, while the detector provides excellent discrimination between 26Al ions and the residual flux of 26Mg ions. This system is now in routine operation for 26Al measurements, and allows us to achieve comparable counting statistics for 26Al and 10Be from the same sample.
Details of the AMS setup, and our experience to date with the methodology will be presented.
See more of New Developments in Terrestrial Cosmogenic Nuclide Research
See more of The 10th International Conference on Accelerator Mass Spectrometry (September 5-10, 2005)