ARGON BEHAVIOR IN GEM-QUALITY ORTHOCLASE FROM MADAGASCAR - EXPERIMENTS AND SOME CONSEQUENCES FOR AR-40 AR-39 GEOCHRONOLOGY/

Citation
No. Arnaud et Sp. Kelley, ARGON BEHAVIOR IN GEM-QUALITY ORTHOCLASE FROM MADAGASCAR - EXPERIMENTS AND SOME CONSEQUENCES FOR AR-40 AR-39 GEOCHRONOLOGY/, Geochimica et cosmochimica acta, 61(15), 1997, pp. 3227-3255
Citations number
36
Categorie Soggetti
Geochemitry & Geophysics
ISSN journal
00167037
Volume
61
Issue
15
Year of publication
1997
Pages
3227 - 3255
Database
ISI
SICI code
0016-7037(1997)61:15<3227:ABIGOF>2.0.ZU;2-8
Abstract
In order to investigate argon diffusion in as simple a K-feldspar stru cture as possible, a single crystal of the gem-quality Itrongay K-feld spar from Madagascar has been studied using cycled step heating, ultra -violet (UV) laser depth profiling, in vacuo crushing, Electron Microp robe Analysis (EMPA) and X-ray diffraction (XRD) techniques. The resul ts have been modelled using both the multidomain model and a multipath model invoking pipe or short circuit (SC) diffusion. Cycle heating (b oth forward and reversed) indicates a very retentive K-feldspar, in wh ich the gnat majority of argon release conforms to a simple model of v olume diffusion with a single domain size, a single activation energy, and is associated with a plateau age of 435 +/- 8 Ma. However, the fi rst 1% of argon release exhibits younger ages and the same type of com plex behaviour seen in other K-feldspars. Modelling the argon release in terms of multidomains yields good fits to data with four domains of varying activation energies. An alternative model for the argon relea se, involving not only volume diffusion through the lattice but also S C or pipe diffusion and mass transfer between lattice and rapid diffus ion paths, provides a good though less sophisticated model to explain the argon release. Argon concentration/depth profiles of a previously outgassed sample, measured using the UV laser ablation technique, exhi bited argon loss in only the upper 5-10 mu m. Argon loss calculated fr om the measured profiles suggests that the low temperature domains out gassed during step heating were within 5-10 mu m of the grain surfaces and thus may be artefacts of sample preparation and surface texture e ffects. In vacuo crushing of an untreated sample released argon with a n older age than the plateau value, which was not detected by any of t he other analytical techniques. However, previously cycled step heated samples contained very little excess argon, suggesting the existence of traps within the feldspar structure into which Ar-40 diffused in na ture to yield apparent excess argon, but which were then filled with c orresponding Ar-39 during cycle heating. The K-feldspar structure illu minated by these studies is one in which argon diffuses through the la ttice over distances of at least 100 mu m probably via a volume diffus ion mechanism to the grain boundaries. However, close to the surface a rgon may also diffuse along fast diffusion paths resulting from natura l traps opened during sample preparation. Though restricted to very sm all volumes of gas in the case of the Itrongay orthoclase gem, the exi stence of the traps has potentially important consequences for Ar-40/A r-39 geochronology applied to K-feldspars. Copyright (C) 1997 Elsevier Science Ltd.