Accueil > EGW 2018 : 6th European Geothermal Workshop > Abstracts > Session 3 : Operating of Geothermal Systems > Session 3 : Poster Presentations

Session 3 : Poster Presentations

S3.1 Coda wave interferometry during the heating of deep geothermal reservoir rocks

Jérôme Azzola, Luke Griffiths, Jean Schmittbuhl, Dimitri Zigone, Frederic Masson, Vincent Magnenet, Patrick Baud, Michael Heap

Coda Wave Interferometry (CWI) is a high-resolution technique that aims at tracking small perturbations in a diffusive medium from the correlation of seismic waveforms (Snieder, 2006). Thanks to the development of seismic noise correlation methods and of the plentiful records of ambient seismic noise in time, CWI has been used in recent years to monitor the fine-scale evolution of fault zones and more recently of deep reservoirs (e.g. Sens-Schönfelder and Wegler (2011), Lehujeur et al. (2014)).

However, to provide a quantitative interpretation of the reservoir monitoring, direct modeling of the physical effects like the influence of the temperature on seismic wave scattering, is required to invert temperature effects from measurements of velocity changes. We quantify here the impact of thermo-elastic deformation on seismic wave diffusion following both a numerical and an experimental approach. The latter is based on experimental results from Griffiths et al. (2018) obtained on a deep geothermal representative granitic rock (i.e. Westerly Granite) for which CWI monitoring is performed during the repeated heating and cooling of the sample. We developed a numerical model that replicates the experiments. The numerical modelling initially proposed by Azzola et al. (2018) combines the thermo-elastic deformation obtained from a finite element approach (Code_Aster) to the wave propagation simulation obtained from a spectral element approach (SPECFEM2D). A stretching technique is applied as in the laboratory to quantify waveform changes. First, the simulations only account for the thermo-elastic dilatation of the bulk and complementary simulations include changes of Young and bulk moduli with temperature. The comparison between the laboratory and simulated results intends to better understand the physical origins of the CWI measurements.

Our study shows first that multiple reflections on the boundaries of the numerical sample reproduce the wave scattering happening in the granite sample that includes a set of mineral inclusions and of micro-cracks. The comparison is based on the wave diffusion model describing similarly both experimental and numerical samples. The samples share also a similar thermo-elastic behavior in the simulation and in the laboratory experiments but only after the second heating and cooling cycle. In addition, we show that the CWI measurements reveal reversible time shifts correlated to the thermo-elastic deformation of the sample in both approaches. Meanwhile, the influence of thermoelastic deformation differs from the numerical proxy to the real granitic sample. We introduce temperature dependence of elastic moduli in the model in order to discuss the role of irreversible deformation (e.g. micro-cracking) in the difference in behavior highlighted. These results suggest that there are open perspectives to monitor thermal strain in geothermal reservoirs using CWI.

S3.2 Inhibitor influence on corrosion of carbon steel with respect to the Upper Rhine Graben geothermal brines

Petra Huttenloch, Roman Zorn, Linda Makni, Hagen Steger, Frank Schilling, Detlev Rettenmaier

Corrosion of construction materials in geothermal brine environments is of great concern in the long-term operation of geothermal power plants. In case of geothermal energy production there are many site-specific factors influencing corrosion processes e.g. high salinity, pH, high temperature, presence of non-condensable gases, fluid velocity, corrosion products, geometry of the power plant, acid treatment and acid cleaning procedures (Guerra & Jacobo 2012). To prolongate the lifetime and thus decrease downtime periods and economic losses, a careful material selection at acceptable costs is essential (Carter & Cramer 1992). Nogara & Zarrouk (2018b) summarized the results of 172 tested metals and alloys for the use in geothermal environment. Among low cost carbon steels, stainless steels, alloys and titanium are employed for functional parts as turbine blade or heat exchanger material respectively (e.g. Ravier et al. 2016). In geothermal power plants up to 12 types of corrosion may occur depending on the corrosiveness of the produced brine (e.g. Nogara & Zarrouk 2018a).

Deep geothermal fluids of the URG have salinities up to 200 g/L and a high CO2 content (e. g. Sanjuan et al. 2016) resulting in corrosion of the construction materials. For example the produced Na-Ca-Cl brine at the geothermal power plant in Soultz-sous-Forêts (Alsace, France) exhibit high salinity with 100 g/L total dissolved solids, a temperature of 160 °C and a gas:liquid volume ration of nearly 1 (mainly CO2) (Sanjuan et al. 2010). Several corrosion studies (e.g. Mundhenk et al. 2014) have been performed in the last years. For the geothermal power plant in Rittershofen material evaluation tests were performed on heat exchanger materials (Ravier et al. 2016).

Corrosion and/or scaling control can help to extend the lifetime of geothermal power plant facilities from downhole to surface like casing, tubings, pumps or heat exchangers. The use of corrosion and/or scale inhibitors is a common method for material protection and one of the most useful in industry due to low cost and good handling (e.g. Finsgar & Jackson, 2014). Inhibitors are mostly organic substances interfering with the anodic or cathodic corrosion reaction, forming a protective barrier on the metal surface. The efficiency of an inhibitor depends on flow patterns, solution chemistry, temperature, pressure, the mode of interaction with the metal surface, specific inhibitor concentration and on the inhibitor properties (Wang et al. 2001). Main aspects influencing inhibitor performance in geothermal environment are : a) the compatibility of the inhibitor in high saline brine (avoiding precipitation, clogging), b) competition of dissolved ions and inhibitor molecules for sorption sites on the metal surface and c) existing scales may have a synergistic or antagonistic effect. However, incorrect choice or use of organic inhibitors can lead to corrosion stimulation. Organic inhibitors have a threshold concentration value, below which no corrosion inhibition occurs (Schweitzer 2010).

The aim of this work is to evaluate the corrosion processes of different steel materials and the performance of several corrosion inhibitors in the geothermal environment of the Upper Rhine Graben. Corrosion behavior of carbon steel in CO2 saturated artificial geothermal brine, was tested in absence and presence of an amine based inhibitor by weight loss tests and electrochemical measurements. Experimental conditions were adapted to the conditions existing at the re-injection side of geothermal power plants located in the Upper Rhine Graben. Experiments were performed with carbon steel in CO2-saturated artificial NaCl- and Na-Ca-Cl brine (pH ≈ 5, 80 °C). The influence of Ca2+-ions on the corrosion behavior due to scale formation and its impact on inhibitor performance was investigated. The inhibitor performance was tested by weight loss tests and the use of electrochemical measurements as potentiodynamic polarization (PP) and impedance spectroscopy (EIS), depending on inhibitor concentration and time. The impact of corrosion scale on inhibitor performance was tested on pre-corroded steel surface as well as the corrosion behavior of blank material in geothermal environment. All experimental tests (weight loss, PP and EIS) showed a decrease in corrosion rate (CR) with increasing exposure time in artificial geothermal brine (Na-Ca-Cl brine, 80 °C, CO2, pH ≈ 5). The formation of a (Fe,Ca)CO3 layer restricts the mass transfer process at the interface metal surface– scale – bulk solution, resulting in lower CR. In the presence of Ca2+-ions the CR decreases compared to pure NaCl-brine at the same chloride content. Both steel qualities exhibit a similar corrosion behavior at same experimental conditions. The experimental tests (weight loss, PP and EIS) showed that the inhibitor performance depends on inhibitor concentration. There is an area of concentration where the inhibitor perform most. Too low and too high inhibitor concentrations were not sufficient. The Presence of Ca2+-ions resulted in the interplay between the formation of carbonate deposits and adsorption of inhibitor molecules on the metal surface resulting in lower inhibitor efficiency at scaling conditions. Accepted CR of carbon steels in geothermal environment < 0.3 mm/a were achieved by inhibitor addition of 200 ppm (Figure 1). Inhibitor performance decreased significantly on pre-corroded metal surface compared to fresh samples at same experimental conditions.

S3.3 Temperature analysis of the well RN15/IDDP2 in Reykjanes under long-term flow variations

J.Wang, M. Gholami Korzani, F. Nitschke, E. Gaucher, T. Kohl

The RN15/IDDP2 well in Reykjanes, Iceland, is one of the demonstration sites of the ongoing EU. Horizon 2020 DEEPEGS project. So far, the well represents the deepest geothermal drilled hole in Iceland with a final depth of 4659 m and measured bottom hole temperature of 427°C and fluid pressure of 34 MPa. (Friðleifsson et al., 2016). The project objective is to explore reservoirs under supercritical conditions. One of the challenging scientific tasks of this project is to estimate the static formation temperature around the well under continuous injection condition within the drilled borehole. Our on-going research aims at applying a numerical simulation approach to inverse the formation temperature based on the recorded temperature logs. Here, we present and discuss our current state of research.

In our work, a transient thermal transport model is set up and calculated by applying an in-house developed wellbore simulator, which allows for the incorporation of complex wellbore configurations and boundary conditions. In addition, the model is capable of taking into account the circulation time variation along depth due to the deepening of the well into account to prevent overestimated-cooling period at the greater depth. As an example test of the numerical tool, we present the procedure of performing one forward modeling using prior known initial formation temperature to obtain well fluid temperature profile. In this process, the whole injection history of drilling mud is considered which allows us to follow the local temperature perturbation that is changing with time at different depths. The time stepping scheme in the simulation is adjusted according to the real logging speed to guarantee simultaneous temporal-depth matching between the simulated and measured fluid temperature. Initial results of the long-term temperature variation in the well as well as the comparison between simulated and logged data will be presented.

11 décembre 2018