METAMORPHIC FORMATION AND STABILITY OF THE CLAY MINERAL SMECTITE IN THE GREENSCHIST ASSEMBLAGE PLAGIOCLASE-CHLORITE-QUARTZ

The occurrence of expandable phyllosilicates in nature as anything other than a product of weathering reactions has seemingly been limited to hydrothermal/low-grade metamorphic reactions. However, experimental studies involving blueschist and greenschist minerals have encountered the persistent appearance of an expandable phyllosilicate at high temperatures (>500°C) and pressures (~0.4-2.5 GPa). These minerals are capable of incorporating variable amounts of water into their structures as is typical of smectite/vermiculite. In this study the greenschist assemblage plagioclase-chlorite-quartz (plg-chl-qtz) was treated at high pressure-temperature conditions to determine whether expandable phyllosilicates would form as a stable phase.

Experiments were done in the system Na₂O-CaO-MgO-Al₂O₃-SiO₂-H₂O using synthetic chlorite, albite (ab), and plag (An₂₅) made at 680°C, 700°C, 700°C, and 0.45GPa, 0.21GPa, 0.39GPa, respectively, and a natural quartz. Treatment at 0.5 GPa of ab-chl-qtz, mixed in the molar ratios of 1:1.4:5, yields smectite starting at 500°C, which grows strongly up to 800°C, along with a talc-like phase (“talc”) and cordierite which both appear at 700°C. Treatment at 0.5 GPa of plag(An₂₅)-chl-qtz, mixed in the molar ratios of 1:1.4:5, has a different sequence of phases with “talc” forming at 600°C, prior to smectite development, which occurs at 700°C and continues to 800°C, with cordierite appearing at 800°C. ”Talc” has the approximate composition Na_.2Ca_.03Mg_2.5(Al_.7,Si_3.8)O₁₀(OH)₂.

The determination of whether or not we were observing stability boundaries in this assemblage lead us to pursue the reversibility of the general reaction plag + chl + qtz + H2O = smectite by using mixtures of both the reactants and products. When attempting to reverse this reaction at 600-700°C and 0.5 GPa for durations of 80-170 hours between plag(An₂₅)-chl-qtz and the smectite-rich reaction products, “talc” becomes the dominant stable phase and the smectite is reacted out. This reaction series proves that smectite is in fact a metastable phase and that the talc-like phase is a stable phase in this assemblage. These results suggest that even though Mg-rich tri-octahedral smectite may nucleate relative to a common greenschist-facies assemblage, the smectite may indeed be metastable relative to "talc".