Lagneau, et al., reported in 2019 that the development of reactive transport soared during the 1990’s, driven by the necessity to demonstrate the long-term efficiency of radioactive waste repositories (Bildstein et al. 2019; Claret, 2019; Cama et al. 2019, both this volume). The approach, based on a rigorous description of processes and their coupling, provides a basis of confidence to bridge the gap in time and space between knowledge gained in laboratory experiments and the dimension and lifetime of a high-level waste repository.
Reactive transport codes progressively increased in complexity, as more processes were included to account e.g., for variably saturated flow, heat transport and more complex chemical processes (Steefel et al. 2015). In the meantime, improved algorithms and computer power opened the way for larger simulations. Reactive transport simulation now embraces a large array of applications in the Earth and environmental sciences. Thus, in the field of contaminant hydrology, such studies address problem
from the interface scale, to quantify migration and retention mechanisms, up to the scale of a watershed integrating multilayer reservoirs and their inherent spatial variability. Over the last decades, extensive monitoring and characterization studies allowed for better calibration of processes, parameters determination and more generally better quantitative understanding of sites (e.g., Hammond and Lichtner 2010; Li et al. 2010; Hammond et al. 2011; Arora et al. 2016) or on less emblematic sites
(e.g., Mangeret et al. 2012; Estublier et al. 2013). As a result, better understanding of mechanisms, quantification of competing retention and migration processes were achieved. They allow for the estimation of plume migration, or the evaluation of remediation strategies.
In these applications, a unique application site is characterized, then a model is built in order to draw conclusions and propose recommendations. Apart from academic or agencies sponsored studies, the industry is also interested by this approach: estimation of reservoir properties in the oil and gas industry (Xiao and Jones 2006), evolution of gas storage sites (e.g., CO2 in Audigane et al. 2007, or H2 in Saínz García et al. 2017), evaluation of impacts of the mining industry (Metschies and Jenk 2011; Molson et al. 2012; Neuner and Fawcett 2015).
Another type of use is also relevant: industrial deployment of reactive transport within operation with a view to support operators in decision-making. Operating conditions are then very different: models should be built on a regular basis on numerous application cases, sometimes relying on incomplete sets of data. Additionally, this type of application only makes sense if it can be integrated in the regular workflow of operations. Hence, timing (to keep pace with the operations), and interfacing (with other aspects of production) are critical. Finally, such deployment can only be performed if the value of the simulations for the operation can be demonstrated.
An illustration of industrial use of reactive transport in operations is given: exploitation of uranium by in situ recovery (ISR). The technique is best suited for large, low grade, medium depth deposits. It consists in dissolving the metal in place by circulating solutions, using a large number of injection and production wells to cover the surface of the field. Relying on the circulation of fluids and their interaction with the deposit, the technique is particularly suitable for reactive transport analysis (Nguyen et al. 1983; Regnault et al. 2012).
Although models can be built to understand the general behavior of such systems, when quantitative assessment of a unit production is wanted, the size of the exploitation and the strong spatial variability inherent to the formations require to build models specific to the local conditions (geology and constraints of exploitation) in each portion of the deposit. After a brief description of the ISR technique, the use of reactive transport simulation is exemplified on the KATCO mine. The modeling approach is presented, including the data necessary to calibrate the model. The conditions to transform a mostly academic application into an operational tool are commented; some examples of use in operation are then proposed.
Finally, in addition to exploitation, reactive transport modeling is also relevant for controlling environmental impacts. Although this application is more traditional, it shows that reactive transport modeling deployed in this context can create the conditions for an integrated approach to better identify post-operational constraints, or even to test alternative exploitation strategies limiting the need for active remediation a posteriori.
The authors concluded that several applications discussed in this paper illustrate how an academic, versatile, reactive transport code like HYTEC can be used in an industrial context. The fast pace imposed by operations, need for quick results, imposed the use of a specific graphical interface, specially devised for this application: HYSR. The code is presently tested, in operation conditions, on the Katco ISR mine. An important lesson from this project is the benefit of co-construction. After the initial
identification of an industrial problem and the possibility to address it with reactive transport tools, a close cooperation was built between an academic team, with the knowledge of the tool and expertise to optimize it, and an industrial team, with the knowledge of industrial context and issues and the driver to improve their simulation capacity. Also, access to an extensive mass of data was key in the early stage of the project to build, calibrate, and optimize the model. As the project moved from R&D department to technical department and finally operations, it became clear the code had to evolve from a versatile, keyword-based input files to a more intuitive interface.
GUI is of course essential, but a key was to calibrate the interface to the needs of the engineers: the GUI is basically devised so that ISR engineers interact with the code using the frame of mind of their trade: inputs organize various data of the project and proposed development scenarios, outputs are displayed following the need of production engineers including their financial conversion for comparison purpose
Managed & Built by I2M Consulting, LLC | © 2010-
