- © The Mineralogical Society Of America
Determination of accurate temperatures for geological events has been the grail of stable isotope geochemistry since the seminal 1947 paper by Urey. In theory, this should be simple. Calibrated mineral pairs are common, analysis is rapid, and there is no significant pressure correction. However, in spite of widespread application, the promise of reliable thermometry has been elusive. Stable isotope temperatures in metamorphic and igneous rocks are often controversial; is the fractionation between two phases a thermometer, speedometer (rate dependant), hygrometer (P(H2O)-dependant), or chimera? A number of factors have contributed to this uncertainty including: incomplete and sometimes conflicting calibration of isotope fractionation factors; limited understanding of the kinetics of mineral exchange; and the lack of microanalytical techniques. This situation has improved dramatically over the past decade, permitting reliable and detailed thermal histories to be inferred.
In a landmark paper, Bottinga and Javoy (1975) empirically calibrated the oxygen isotope fractionation factors among ten rock-forming minerals at T > 500°C. Their compilation shows that 84% of analyzed samples contain at least three minerals meeting a concordance test for equilibrium. They concluded “the great majority of igneous and metamorphic rocks …. has conserved a state of oxygen isotope exchange equilibrium.” If correct, this bodes well for thermometry. However, this conclusion was contested by Deines (1977) who critically examined the existing data and concluded that “less than half of the rocks analyzed to date would yield concordant 18O-derived temperatures.” While seeming irreconcilable, both conclusions are based on fact. All rocks contain some degree of disequilibrium, but many also preserve a thermometric record. It is the purpose of this review to discuss the new strategies for reliable isotope thermometry and to explore disequilibrium processes so they may be recognized and properly interpreted.
The organization of this paper will be to discuss the assumptions …