Géosciences Montpellier

• Planète et Univers/Sciences de la Terre/Géophysique
  
• Sciences de l'environnement/Milieux et Changements globaux
  
• sdu/sdu.stu/sdu.stu.hy
  
• Planète et Univers/Sciences de la Terre/Géochimie
  
• Physique/Physique/Géophysique
  
• Planète et Univers/Sciences de la Terre/Pétrographie
  
• Planète et Univers/Sciences de la Terre
  
• Planète et Univers/Sciences de la Terre/Minéralogie
  
• Planète et Univers/Interfaces continentales, environnement
  

Results of reactive transport laboratory experiments often suggest that pore scale heterogeneity induces localization of reactions (the generation of local micro environments favoring reactions that would not occur in a well-mixed Representative Elementary Volume, REV). Multi-Rate Mass Transfer (MRMT), which has been employed to reproduce hydrodynamic heterogeneity, may also be used to simulate geochemical localization. We extended the Water Mixing Approach (WMA) designed for single porosity media, to simulate chemical reactions caused by the mixing of mobile and immobile zones. The method is termed Multi-Rate Water Mixing (MRWM). The MRWM approach was employed to simulate laboratory experiments of CO2-rich brine transport through carbonate rich samples (Luquot et al., 2016). Chemical heterogeneity in space was reproduced by varying the mineral assemblages in immobile regions. This enabled us to reproduce the generally low pH environment while allowing for high pH local zones required for the localized precipitation of kaolinite, which has been observed in reality, but cannot be modeled with conventional reactive transport formulations. The resulting model is very rich, in that it can reproduce a broad range of pore scale processes in a Darcy scale model, and complex, in that the interaction between chemical kinetics and immobile zones physical parameters is non-trivial.

This paper provides an overview of the site characterization work at the Heletz site, in preparation to scientifically motivated CO2 injection experiments. The outcomes are geological and hydrogeological models with associated medium properties and baseline conditions. The work has consisted on first re-analyzing the existing data base from ∼40 wells from the previous oil exploration studies, based on which a 3-dimensional structural model was constructed along with first estimates of the properties. The CO2 injection site is located on the saline edges of the Heletz depleted oil field. Two new deep (>1600 m) wells were drilled within the injection site and from these wells a detailed characterization program was carried out, including coring, core analyses, fluid sampling, geophysical logging, seismic survey, in situ hydraulic testing and measurement of the baseline pressure and temperature. The results are presented and discussed in terms of characteristics of the reservoir and cap-rock, the mineralogy, water composition and other baseline conditions, porosity, permeability, capillary pressure and relative permeability. Special emphasis is given to petrophysical properties of the reservoir and the seal, such as comparing the estimates determined by different methods, looking at their geostatistical distributions as well as changes in them when exposed to CO2.

Heletz sandstone (Israel) reactivity is studied by means of CO2-rich brine percolation experiments and modeling. Experiments are performed at in situ temperature and pressure conditions with two distinctly different injection flow rates and for two different brine compositions: the genuine Heletz brine and a similar brine but in equilibrium with gypsum. The Heletz sandstone is composed of quartz, dolomite, ankerite, K-feldspar, clay minerals and minor amounts of gypsum and pyrite. The main reaction is the dissolution of the carbonates which is fastest for the highest flow rate injection and for the genuine Heletz brine. Permeability increases during the five percolation experiments, regardless of flow rate and brine type. This increase in permeability is associated to a decrease in porosity during the gypsum-equilibrated brine experiments due to gypsum precipitation. The precipitation of secondary minerals (kaolinite, muscovite and smectite) is mainly controlled by the dissolution of K-feldspar and largest at low flow rate injection, indicating that these precipitation processes are diffusion-controlled and localized in immobile zones where favorable chemical micro-environments exist. These processes are quantified by means of a reactive transport model. These results indicated that the CO2 injection at the Heletz site should trigger both permeability increase in interface between brine and CO2, which may extend over a large extension and where mixing processes are very active, especially near the well under intermittent injection conditions.

The main goal of the monitoring stage of Carbon Capture and Storage (CCS) is to obtain an accurate estimation of the subsurface CO2 accumulation and to detect any possible leakage. Laboratory experiments are necessary to investigate the small scale processes governing the CO2–brine–rock interaction. They also provide a means to calibrate the results coming from field scale geophysical methods. In this work we set up an experimental system which is able to perform Electrical Resistivity Tomography (ERT) measurements on centimeter-scale rock samples at various P–T conditions. We present the results of two new experiments related to CO2 monitoring, performed on a cylindrical (4 × 8 cm) Fontainebleau rock sample. In the first one, we have quantified the CO2 saturation at different volume fractions, representing zones from a deep saline aquifer with varying degrees of saturation. In the second one, we have monitored and quantified the effect of CO2 dissolution in the brine at a pressure of 40 bar during eight days, emulating the invasion of CO2 into a shallow aquifer. Results highlight the importance of accounting for the contribution of surface conductivity in highly CO2-saturated regions, even in clay-free rocks, and also for brine conductivity variation due to CO2 dissolution. Ignoring any of these effects will end up in a CO2 saturation underestimation. We present a modified CO2 saturation equation to account for these two influences.

A shallow field experimental site for CO2 injection was established at Maguelone (Languedoc, France), in order to test in an integrated manner a suite of surface and downhole hydrogeophysical monitoring methods. The objective is to improve monitoring of gas transport in the shallow subsurface and to determine the sensitivity of CO2 monitoring systems for leakage detection. The site offers a natural laboratory to study the processes associated with CO2 injection in a clastic and clay-rich context saturated with saline fluids. Prior to CO2 injection, three nitrogen (N2) injections were undertaken in 2012 to measure the site response to neutral gas injection. In 2013, a volume of 111 m3 (mass of 220 kg) of CO2 was injected during 3.5 h at 15 m depth. During each experiment, the gas plumes were successfully detected from pressure monitoring, time-lapse induction logging and downhole resistivity monitoring with dipole–dipole array. Increases in resistivity are attributed to free gas propagation (either N2 or CO2) whereas decreases in resistivity correlate with CO2 dissolution in the pore fluid. Chemical analyses confirm this hypothesis with a decrease in pH and an increase in the concentration of dissolved species in the latter case.