太古宙花岗质岩石 Si-O 同位素对岩石成因和板块构造的约束

The operation of plate tectonics makes planet Earth unique within our solar system. However, 'When did plate tectonics begin on Earth?' remains hotly debated in the field of geology. Subduction is one of the most distinct signatures of modern plate tectonics. Thus, a critical aspect in addressing this question is to investigate when supracrustal material recycling appeared. Paired Si-O isotopes have recently been identified as powerful paired tracers of surficial materials in the Archean. However, due to the highly metamorphosed rock affinity in most Archean rocks, Si isotopes of these rocks need further evaluations on whether they represent the original Si signature. Besides, Si isotopes display very limited variation during high-temperature magmatic differentiation, which requires ultra-high analytical precision. Otherwise, the data with large analytical errors may hide any potential shifts. Here, we review the progress of the Si-O isotope studies on Archean granitoids (mostly tonalite-trondhjemite-granodiorite, TTGs), and discuss the implications and limitations of the current studies. First, we review the principles of Si isotope application on TTGs and the advantages of paired Si-O isotopes. Second, we discuss the potential problems of the pioneering studies and propose a specific evaluation process. Third, we compile the available global Si-O isotope data, Ge/Si ratios, and other geochemical and geological evidences, and confirm that supracrustal material recycling first appeared at ca. 3.8Ga. Finally, we discuss the remaining issues and provide specific topics for future research.