Addressing Water and Energy Challenges with Reactive Transport Modeling

Perspectives

Sustainable supply of water and energy is one of the grand challenges and opportunities facing the environmental engineering community for the 21st century. Designing and implementing impactful solutions require creative and sustainable use of the subsurface and other Earth's natural environments, as well as technological innovations in treatment and other engineered systems (Fig. 1). Fundamental and applied research that support the development of such solutions need to capture complex system dynamics, which makes reactive transport modeling that is capable of simulating multiphysics, for example, coupled biological–chemical–physical processes, across scales extremely valuable.

OPEN IN VIEWER

FIG. 1.

Conceptual illustration of Earth's critical zones and subsurface and engineered systems for addressing our society's water and energy challenges. Reactive transport modeling is playing a critical role in related emerging research areas, including recovery and valorization of critical resources and materials, groundwater resource management for drought mitigation, negative carbon emissions, and subsurface renewable energy. The inset figures highlight a multiphase environment at pore scale that can be encountered in many natural systems and a membrane system as an example of the engineered systems.

Reactive transport models (RTMs) started with a handshake between traditional groundwater modeling and geochemical modeling, and embody the coupling of groundwater flow and solute transport equations with the laws of mass action and kinetic rate equations to predict the production, fate, and transport of reactive species. The origins of groundwater modeling focused on water resources, and was extended to include transport of simple solutes (Kitanidis and Bras, 1980; Celia et al., 1989). Geochemical models were developed to apply the principles of chemical thermodynamics and kinetics to calculate speciation at equilibrium or track the temporal reaction path in complex chemical systems (Helgeson, 1968). Initially limited to batch systems, geochemical models were used to investigate reaction networks and chemical evolution in Earth and environmental systems such as groundwater aquifers (Plummer et al., 1990; Postma et al., 1991; Zhu, 2012).

Eventually science drove the need to track the fate and transport of complex emergent contaminants that could not be represented by simple reactions such as linear partitioning and exponential decay. Furthermore, the need emerged to account for contaminants that exist in different oxidation states, different aqueous species, and multiple phases (e.g., non-aqueous phase liquids, water), depending on variations in environmental conditions. This scientific need for models that couple flow and transport with biogeochemical speciation and reactions co-occurred with advancements in computational power, data storage, and unprecedented accessibility to numerical simulation tools. This prompted a surge in the development of RTMs in the 1990s (Yeh and Tripathi, 1991; Lichtner, 1996; Steefel and MacQuarrie, 1996; Clement et al., 1998; Yabusaki et al., 1998; Sun and Clement, 1999; Xu et al., 1999).

In the early 2000s, reactive transport modeling grew into a mature and well-established research field, and has become an important tool for investigating both the subsurface and Earth's critical zones (Steefel et al., 2005; Li et al., 2017) (Fig. 1). Modern RTMs commonly implement a full (bio)geochemical framework that considers chemical heterogeneity, multiple organic and inorganic chemicals, and complex parallel and consecutive reaction pathways. Rich data sets of thermodynamic properties, kinetic mechanisms and rate constants, and distributions of mineralogy and surface area at small scales, among others, generated through advances in molecular level modeling and probing techniques (e.g., microscopy and spectroscopy) are now also incorporated into RTMs (Tokunaga et al., 1998; Goldberg et al., 2007; Scheibe et al., 2009; Y. Wu et al., 2011; Viswanathan et al., 2012; Deng et al., 2016, 2020; Beckingham et al., 2017).

RTM has a long history of application in environmental engineering and science, which is steadily increasing (Fig. 2). Over the 30 years between 1990 and 2019, 33% of publications from a Web of Science search of the keyword “reactive transport model*” are tagged with Environmental Sciences, and 14% categorized as Environmental Engineering, ranking #1 and #5 among all disciplines. Water Resources has consistently been the primary application area for RTM, and in the past decade, energy emerged as one of the primary application areas.

OPEN IN VIEWER

FIG. 2.

Number of records from “reactive transport model*” keyword search in the Web of Science (© 2020 Clarivate Analytics) for the periods of 1990–1999, 2000–2009, and 2010–2019. The pie charts highlight the relative percentages of records associated with specific subject areas. The top 10 subject areas in the period of 1990–2019 are shown in discrete colors, including Environmental Sciences (Env. Sci.), Water Resources (Water Res.), Multidisciplinary Geosciences (MD Geo.), Geochemistry Geophysics (GC&GP), Environmental Engineering (Env. Eng.), Limnology, Energy & Fuels, Computer Science Interdisciplinary Applications (Comp. Sci.), Chemical Engineering (Chem. Eng.), Civil Engineering (Civil Eng.). All other subject areas are combined in “All others.” Note that a record might be associated with more than one subject area, and some of these subject areas emerged after 2000 or 2010.

Overall, the top three application areas of reactive transport modeling that are of particular interest to the environmental engineering and science community in the past three decades are groundwater contamination and remediation (Walter et al., 1994; Mayer et al., 2001; Zhu et al., 2001; Gibson et al., 2011; Wallis et al., 2011; Yabusaki et al., 2011; Prommer et al., 2019), mine waste pollution management (Wunderly et al., 1996; Runkel and Kimball, 2002; Amos et al., 2004; Jurjovec et al., 2004; Mayer et al., 2015), and subsurface systems such as nuclear waste disposal and geological carbon storage (Spycher et al., 2003; Goldberg et al., 2007; Xu et al., 2010; Aradottir et al., 2012; Viswanathan et al., 2012; Navarre-Sitchler et al., 2013). Patil and McPherson (2021) in this special issue presents a review of reactive transport modeling in fault zones, which is an important feature that controls solute transport and fluid migration in the underground. It includes a thorough discussion of different conceptual models and numerical approaches, with a focus on self-sealing and self-enhancing of the fault zones as a result of water–rock interactions.

Natural and engineered systems are rarely static, and are subject to alteration due to processes such as mineral dissolution and precipitation and biofilm growth. Current RTMs can account for dynamic changes in physical properties of the systems investigated by tracking porosity and surface area evolution in continuum-scale models (Xie et al., 2015) and mineral–fluid interface geometry in pore-scale models (Molins et al., 2020). These expanded capabilities make long-term predictions about system evolution that draw inferences from much shorter-term laboratory experiments possible with RTMs. This development of dynamic modeling capability has benefited enormously from the development of nonintrusive imaging techniques and 4D scanning (Deng et al., 2015, 2020; Voltolini and Ajo-Franklin, 2020). For instance, the RTM developed by Ling et al. (2021) in this special issue is based on x-ray micro-computed tomography imaging, synchrotron micro x-ray fluoresence, and micro x-ray diffraction data from experimental studies.

As the field of RTM moves forward, the multitude of advances provide the backbone for further development of multiphysics codes. Existing RTMs can be readily expanded to couple additional physical processes (Beisman et al., 2015; Tournassat and Steefel, 2019), which will broaden the role of RTMs in environmental studies. For instance, recent developments include models that solve energy balance across the land–atmosphere interface and surface flow equations to expand application of RTM to coupled surface water–groundwater systems and consider climate forcings (Beisman et al., 2015; Li et al., 2017). Advances in subsurface RTM include the coupling between reactive transport and mechanical processes (Spokas et al., 2019). As shown in Gonzalez-Estrella et al. (2021) in this special issue, geochemical reactions between brine and cement and the confining stress acted together in controlling the hydraulic and mechanical properties of fractured cement. Both mineral precipitation and deformation of fracture asperities contributed to a decrease in fracture hydraulic conductivity. Further development of RTMs to account for such interactions is required for more accurate assessments of wellbore integrity, cement-based or other materials used for resilient infrastructures (e.g., permeable pavement systems and low-carbon cements).

RTMs can also be readily coupled with statistical methods or machine learning algorithms for sensitivity analyses and to quantify uncertainties associated with thermodynamic and kinetic data and other model inputs. It cannot be overemphasized that this type of coupling can only be meaningful when the RTMs are tested and validated against experimental measurements and/or field observations. Li et al. (2021) in this special issue reports a global sensitivity analysis of the controlling parameters for scale development in shale-hydraulic fracturing fluids systems. A validated RTM was used to generate data under a multitude of conditions to systematically evaluate the key controls and to inform operations that minimize the environmental impact and water consumption. These advancements will bring an expansion of the use of RTM to inform policy making. RTM results with quantifiable uncertainty can be incorporated in risk analysis (Siirila et al., 2012) and RTMs can be directly included as a module in integrated assessments. Deng et al. (2017) is among the first to present a framework that couples process-based physical modeling and integrated assessment models to investigate leakage risks associated with geological carbon storage and the impacts on global energy systems and carbon mitigation efforts. This kind of framework can be leveraged to expand the role of RTMs in policy making.

Emerging RTM Applications

After continuous advancements, RTMs are expected to play an increasing role in emerging areas that are of interest to environmental engineering and science communities and that are critical for addressing water-energy challenges our society faces (Clarens and Peters, 2016; National Academy of Engineering and National Academies of Sciences, Engineering, and Medicine, 2019). In this study, we highlight a few example application areas (as also illustrated in Fig. 1), which have demonstrated great potential for leveraging/developing RTM capabilities to examine system-specific processes and parameters to inform reverse engineering and operation optimization.