U-Pb geochronology of grossular-andradite garnet

Garnet has the unparalleled ability to record both the conditions and the timing of its growth. The major and trace elemental as well as stable isotopic composition of garnet track changing P-T and fluid conditions throughout growth (Jamtveit et al., 1993, Crowe et al., 2001, D'Errico et al., 2012, Zhai et al., 2014). These conditions can at present be directly linked to absolute timing using Sm-Nd or Lu-Hf geochronology (e.g., Duchêne et al. (1997); Caddick et al. (2010)). Recent analytical advances, including micro-sampling and partial dissolution techniques, have enhanced both the spatial and temporal resolution of Sm-Nd geochronology and its ability to recover detailed P-T-t histories (Harvey and Baxter, 2009, Pollington and Baxter, 2011). Compared to the Sm-Nd system, U-Pb geochronology has a number of distinct advantages. It can be performed using in-situ methods, such as LA-ICPMS or SIMS, and it does not require complementary co-genetic phase or whole rock measurements to calculate an age. This work presents a method to apply U-Pb geochronology, using LA-ICPMS analysis, to grossular-andradite series (Ca₃Al₂Si₃O₁₂–Ca₃Fe₂Si₃O₁₂) garnets.

The grossular-andradite solid solution series commonly occurs in calcsilicate rocks formed in skarn-type contact metamorphic and hydrothermal environments. From an economic standpoint, many important metals, including Fe, Cu, W, Zn, Mo, etc. are sourced from mineralized skarn deposits (Meinert, 1992). Also, recent studies argue that the formation of skarns may act as a major flux of atmospheric CO₂ throughout geologic history (Lee et al., 2013, Lee and Lackey, 2015). The magmatic-hydrothermal systems which form skarn deposits are inherently short-lived, with geochronologic data suggesting lifespans up to a few million years, while numerical models predict timescales of only tens of thousands of years (Maksaev et al., 2004, Deckart et al., 2005, Chiaradia et al., 2013, Chelle-Michou et al., 2015). Metasomatic processes, such as skarn formation, can be indirectly dated by zircon U-Pb geochronology of the causative pluton or cross-cutting dikes (Meinert, 1992, Chiaradia et al., 2013, Yao et al., 2014). Direct dating of magmatic-hydrothermal processes is limited to accessory minerals, including Re-Os of molybdenite, U-Pb of titanite, or Rb-Sr and ⁴⁰Ar/³⁹Ar of micas (Maksaev et al., 2004, Deckart et al., 2005, Chiaradia et al., 2013, Chelle-Michou et al., 2015). These phases are not ubiquitous and the timing of their formation in the overall history of hydrothermal activity is often unclear. Grossular-andradite, however, is a primary product of metasomatism and contains a record of fluid processes in its major, trace, and stable isotope zonation (Jamtveit et al., 1993, Crowe et al., 2001, D'Errico et al., 2012, Zhai et al., 2014). The ability to directly date skarn grandite has the potential to decipher how hydrothermal systems develop through time.

The garnet mineral system has presented a challenge for U-Pb geochronology, in part because almandine, pyrope, and spessartine-rich garnets typically contain less than ~ 100 ppb U or Th (Haack and Gramse, 1972, Guo et al., 2016). Pioneering garnet U-Pb geochronology on almandine and pyrope-rich garnets from regional metamorphic environments using ID-TIMS proved consistent with other geo/thermochronometers (Mezger et al., 1989 and Mezger et al., 1991), but measurable U and Th within Fe-Mg garnets is likely sourced from inclusions (DeWolf et al., 1996). Grandite-series garnets, on the other hand, contain ppm levels of U and Th (Haack and Gramse, 1972, Lal et al., 1976, DeWolf et al., 1996, Yudintsev et al., 2002). DeWolf et al. (1996) and Meinert et al. (2001) first demonstrated the potential of U-Pb geochronology of grossular-andradite garnet using ID-TIMS methods.

This study uses high-accuracy ID-TIMS U-Pb geochronology to demonstrate U-Pb age homogeneity and reproducibility at the ca. 1% level and to establish a viable primary reference material. We then develop grandite LA-ICPMS U-Pb analytical protocols and demonstrate their efficacy on two other potential reference materials of varying age and composition. The successful methodology is applied to a case study of a Late Miocene skarn system on Serifos Island, Greece associated with a complex intrusive history.