Two case studies of fractured upper mantle peridotite from the Western Gneiss Region (Norway) and Lherz (French Pyrenees): did reaction-induced fracturing play a role?
MetadataShow full item record
The possibility of in situ CO2 sequestration in mantle rock is widely investigated (Kelemen & Matter (2008) amongst others). The process involves infiltration of CO2-rich fluids into the mantle rock through initial cracks, which react with the main constituents of the host rock (olivine and pyroxene) to form serpentine and carbonates. Hereby the CO2 content in the atmosphere is reduced. A volume increase during both serpentinization and carbonation could cause further fracturing of the rock. In this case, the fractures would enhance the process by creating additional reaction surface and allowing fluids to penetrate deeper into the rock. Studying naturally fractured mantle rocks gives insight in the mechanisms that play a role in the formation of fracture networks and fracture propagation. Two case-studies of fractured peridotite bodies in the Western Gneiss Region (Norway) and the brecciated Lherz body (French Pyrenees) were done. The development of fracture networks and the role of serpentinization and carbonation in fracturing were investigated by studying the geometry and scale of fracture patterns in the field and performing optical and electron microprobe (EMP) analyses on thin sections. The peridotite bodies of Raudhaugene and Ugelvik in Norway are characterized by a variety of fracture networks at different scales. A homogeneous serpentine network is observed at micro-scale. At outcrop scale, fracture networks consist of thin (1 mm) extensional fractures filled with serpentine. Fractures cut each other and the compositional banding fairly randomly, although fractures at angles of 60, 90 and 120 degrees also occur. At some localities thick serpentine veins with a second generation of perpendicular fractures were found and parallel talc veins have been observed. Pyroxenite layers are characterized by a regular fracture pattern of equally spaced fractures (sub)perpendicular to the pyroxenite layer. The fractures show a dominant orientation. Crack-seal vein microstructures are observed within the fractures. The homogeneous serpentine network indicates widespread fluid infiltration, most likely via a micro-fracture network that formed in an earlier stage. As there is no gradient in the degree of serpentinization throughout the peridotite body, the micro-fracture network is not likely to be the result of reaction-induced fracturing. It is more plausible that the micro-scale fracture network has formed due to thermal cracking. The crack-seal structure observed in the fractures in the pyroxenite layers shows that extension has taken place during serpentinization. The regularly spaced fractures are partly explained by differential volume expansion due to serpentinization. The predicted volume change in the adjacent peridotite is large enough to cause the observed extension in the pyroxenite layers. The dominant orientation of the fractures in the pyroxenite layers is explained by influences from regional stress. Outcrop-scale fractures in the peridotite do not show evidence for reaction-induced fracturing. More likely they formed due to multiple phases of regional stress and hierarchical fracturing. Different breccias are recognized in the Lherz peridotite body, which are characterized by angular or rounded peridotite clasts, a calcite or peridotite matrix, and the presence of exotic clasts. Cross-cutting relations reveal that serpentinization occurred before fracturing and the formation of carbonate veins. Rotated angular clasts and calcite geodes cannot be explained by reaction-induced fracturing, rather they have a sedimentary origin. This is confirmed by lamination in the matrix consisting of calcite and peridotite particles. Rounded clasts and the incorporation of exotic clasts in the breccia also indicate a sedimentary setting. The Lherz body was most likely intensely fractured due to tectonic stress and subsequently reworked by sedimentary processes. The absence of Mg-rich carbonates in both geological settings, indicate that in situ carbonation has not taken place. Carbonates are present in the form of calcite in fractures, but their origin is most likely outside the system and no evidence has been found that they are involved in fracturing. In this study insights were gained in fracture patterns and various mechanisms of fracture development. Clear evidence for reaction-induced fracturing, resulting in peridotite fracture networks, was not observed.