K content of the K-feldspars being measured in optical dating or in thermoluminescence dating

Abstract

Mejdahl (1983, 1985) suggested the use of separated potassium feldspars for dating, and showed the feasibility of this. We address here the question of what is the correct potassium content to use to evaluate the β dose rate from the potassium within the grains. Pure K-feldspar, KAlSi3O8, has 14.0 wt. % K. The grains in a sediment generally consist of a highly inhomogeneous collection and one can expect grains to be present with a range of K contents between 0 and 14%. The laboratory procedures designed to separate Kfeldspars are not perfect and usually result in the presence of other minerals such as quartz and plagioclase feldspars. The situation is further complicated by the fact that different mineral grains have different luminescence sensitivities and it is not unusual to find that most of the light that is being measured arises from a small fraction of the grains of an aliquot. The one observation that appears to make the problem tractable is that it is generally observed that there is a positive correlation between luminescence sensitivity and potassium content. Our approach to the problem has been to note that 14% is the maximum possible value of the K content, and that a minimum value can be determined from the average K content of the separated grains. The grains are separated using standard procedures. These are an HCl treatment to dissolve carbonates, sieving to select a grain size, a 6 minute HF etch to clean, but not dissolve the grains, an HCl treatment to dissolve precipitated fluorides, sieving to remove small fragments, selection of the < 2.58 g cm fraction using an aqueous solution of sodium polytungstate, and magnetic separation. After optical dating measurements the grains are recovered and about 0.3 g sent for commercial atomic-absorption analysis; the K contents so obtained for a variety of samples are given in the table.