Investigations of radiogenic isotope geochronology and element zoning in garnets: Partial dissolution studies, implications for geochronology, and a case study from the Zermatt-Saas ophiolite, Western Alps

Continent-continent collisions are fundamentally driven by processes that occur in the deep crust and upper mantle, far from direct observation. Recognition of this has led to interest in terranes that have been subjected to high-pressure (HP) metamorphism because of the geodynamic problems associated with subduction of crust to deep levels, followed by rapid exhumation required to preserve HP assemblages. The most commonly used long-lived radioactive nuclides of the rare-earth elements, 147Sm and 176Lu, are well-suited for geochronology of prograde metamorphism of garnet-bearing lithologies. The distinct chemical behavior of Lu and daughter-product Hf can limit complete sample dissolution, however. Tests indicate that fractionation of Lu from Hf occurs during leaching processes and/or partial dissolution. Although acid-leaching procedures may improve Sm-Nd geochronology, these procedures succeed for this isotope system only because of their chemical similarity, whereas the distinct chemical behavior of Lu and Hf during acid-leaching or selective dissolution may produce inaccurate Lu-Hf isochrons. High-temperature and pressure dissolution such as that which can be achieved through Parr-bomb dissolution is critical to ensure complete dissolution of refractory phases such as garnet or zircon.; Five new Lu-Hf ages average ∼49 Ma from three localities of the Zermatt-Sacs ophiolite (ZSO); growth modeling and Lu zoning from these samples suggests garnet growth began as early as 90 m.y. Samples from a different locality, however, produced ages of ∼40 Ma, and growth models suggest initiation of garnet growth occurred at ∼60 Ma; this timing suggests a later prograde history for the ZSO supporting the idea that the ZSO was subducted diachronously. The protracted duration of garnet growth (∼20 to 50 m.y.) required to accommodate the zoning and geochronology modeling data here suggests the ZSO was partially coupled to continental crustal rocks during its prograde history, which most likely caused slow subduction. The proximity of the ZSO in the Saas-Fee region to the continental Sesia and Monte Rosa units and the similarity of ages suggest these units may have been linked for part of the ZSO tectonic history and that the Monte Rosa may be partly responsible for the rapid exhumation of portions of the ZSO.