GNGTS 2019 - Atti del 38° Convegno Nazionale

206 GNGTS 2019 S essione 1.3 TOWARDS A UNIFIED MODEL OF THE EFFECT OF TEMPERATURE ON PHYSICAL PROPERTIES OF CARBONATIC ROCKS: IMPLICATIONS FOR VOLCANO DEFORMATION S. Vinciguerra 1 , F. Vagnon 1 , C. Comina 1 , A.M. Ferrero 1 , G. Mandrone 1 , R. Missagia 2 1 Department of Earth Sciences, University of Turin, Turin, Italy 2 D.Sc. Eng. de Reservatório e Exploração LENEP/UENF, Macae’, Brasil The physical and mechanical behaviour of carbonate rocks is influenced by five main factors: mineralogy, structure, temperature, stress and time. The effect of temperature on carbonate rocks has been considered only in recent years, since volcano deformation, high temperature gradients in the upper crust and rock-engineering applications, such as drilling, deep petroleum boring, geothermal energy extraction, nuclear waste disposal, CO 2 sequestration, ornamental purposes, etc., require its study. In detail, in volcanic areas the presence of large sub-volcanic carbonate sedimentary successions within volcanic systems is common. Carbonate sequences are potentially prone to thermally-induced reactions. Heat, provided by magmatic activity (Heap et al. , 2013 and references therein), can result in detrimental mineralogical, chemical, and textural modifications to carbonate rock, leaving it intensely altered, fractured, and thus weakened (Heap et al. , 2013 and references therein). This degradation process is triggered by decarbonation and implies the decomposition of calcite (CaCO3) into lime (CaO) and carbon dioxide (CO 2 ) (Mollo et al. , 2012 and references therein). These debilitating chemical changes had a dramatic influence on the physical properties of carbonates. Thermal effects induce increase of porosity and decrease of P- and S-wave velocities and bulk sample density. The strength of the carbonates is dramatically reduced at high in-situ temperatures. Mollo et al. (2012) studied thermal decomposition within a closed system (i.e. no free outflow of CO 2 ) by controlling the carbon-dioxide fugacity during experiments. They found that decarbonation is arrested in a closed system and argued that closed systems might not be realistic due to thermal microcracking, whereby any CO 2 produced would likely escape through the enhanced permeability due to the microcracks. However, these experiments were conducted at ambient pressure (an effective pressure of zero) in the CO 2 fugacity experiments. Therefore, these results may not reflect the process at depth where confining pressure, temperature and differential stresses would influence the physical and mechanical behavior of the rock mass. When processes at depth are simulated (increase in confining pressure), combined with high temperatures representative of volcanic systems, the mineral composition of carbonates (i.e. no presence of lime instead of calcite) does not significantly change on a laboratory time scale. This can be due to a sufficiently lowered permeability (Bakker et al. , 2015) and suggests that ductile deformation is dominant at volcanic temperatures and high confining pressures representative for a volcanic edifice (Bakker et al. , 2015). In contrast, the lowered rock strength due to decarbonation reactions suggested by Heap et al. (2013) is effective in an open system at shallow levels such that CO 2 is able to diffuse slowly over geological timescales. In deeper parts of the system, the permeability of pure calcite limestone will have significantly decreased such that decarbonation reactions are considerably limited, suggesting that ductile deformation is likely to be pervasive and widely distributed within sedimentary basements at depths of 4–5km. Also the expelled carbon dioxide can also potentially enrich nearby magmatic bodies (e.g. Chiodini et al. , 2011) contributing to the overall CO 2 decarbonation, which is considered a reliable marker of impending eruptions (e.g. Aiuppa et al. , 2006). Furthermore CO 2 can build up directly as local pore pressure, changing the effective confining pressure and subsequently affect the mechanical behavior of the sample. For example, as the brittle ductile transition is pressure dependent, a reduction in effective confining pressure could cause a switch from