The effect of chlorite interlayering on 40Ar?39Ar biotite dating: an 40Ar?39Ar laser-probe and TEM investigations of variably chloritised biotites

Abstract

Biotite is one of the most common minerals dated by the 40Ar–39Ar method. It frequently shows K contents below the expected stoichiometric value, suggesting the presence of low-K impurities. The most common low-K alteration product of biotite is chlorite. Therefore, it is important to understand the effects of chlorite interlayering on 40Ar–39Ar ages in order to correctly interpret 40Ar–39Ar data. This study examines the outcome of 40Ar–39Ar dating analyses on variably chloritised biotites from Ordovician intrusive rocks. The infrared (IR) laser-probe technique and different gas extraction methods were adopted. Incremental laser-heating data on bulk samples yielded hump-shaped age profiles with meaningless young and old age steps. Both the extent of anomalous old age steps and the degree of discordance of the age spectra were much more pronounced in the more chloritised biotite samples. In contrast, in situ data on rock chips and total-fusion ages on single biotite flakes yielded ages concordant with, or younger than, the inferred emplacement ages. Transmission electron microscopy (TEM) was used to texturally characterise biotite samples at the nanometre scale. It was also used to document the complex decomposition-transformation process affecting interlayered biotite–chlorite during in-vacuo IR-laser heating to temperatures ranging from ~600 to >1,000 °C. TEM results suggest that hump-shaped age profiles result from an interplay between 39ArK redistribution by recoil during sample irradiation and differential release of argon isotopes hosted in three main reservoirs. These reservoirs are (from least to most retentive): extended defects, chlorite and biotite. The final descending age segment is attributed to the progressive release of argon with increasing temperature from large biotite domains for which 39ArK recoil loss was less important. 40Ar–39Ar data support previous findings, which suggest that 40Ar–39Ar ages when recoil effects are minimised, provide minimum estimates that approach the true biotite age, when the pristine domains are analysed. The most effective approach for obtaining meaningful 40Ar–39Ar ages was using individual total-fusion analyses on carefully selected, single flakes previously split along the basal cleavage by wet-grinding and corresponding to a sample mass of a few micrograms.