Search NASA⌕ Search

SEARCH · Search NASA

Results for “Mesh Optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33

Numerical Investigation of a Gasoline-Like Fuel in a Heavy-Duty Compression Ignition Engine Using Global Sensitivity Analysis

Fuels in the gasoline auto-ignition range (Research Octane Number (RON) > 60) have been demonstrated to be effective alternatives to diesel fuel in compression ignition engines. Such fuels allow more time for mixing with oxygen before combustion starts, owing to longer ignition delay. Moreover, by controlling fuel injection timing, it can be ensured that the in-cylinder mixture is “premixed enough” before combustion occurs to prevent soot formation while remaining “sufficiently inhomogeneous” in order to avoid excessive heat release rates. Gasoline compression ignition (GCI) has the potential to offer diesel-like efficiency at a lower cost and can be achieved with fuels such as low-octane straight run gasoline which require significantly less processing in the refinery compared to today’s fuels. To aid the design and optimization of a compression ignition (CI) combustion system using such fuels, a global sensitivity analysis (GSA) was conducted to understand the relative influence of various design parameters on efficiency, emissions and heat release rate. The design parameters included injection strategies, exhaust gas recirculation (EGR) fraction, temperature and pressure at intake valve closure and injector configuration. These were varied simultaneously to achieve various targets of ignition timing, combustion phasing, overall burn duration, emissions, fuel consumption, peak cylinder pressure and maximum pressure rise rate. The baseline case was a three-dimensional closed-cycle computational fluid dynamics (CFD) simulation with a sector mesh at medium load conditions. Eleven design parameters were considered and ranges of variation were prescribed to each of these. These input variables were perturbed in their respective ranges using the Monte Carlo (MC) method to generate a set of 256 CFD simulations and the targets were calculated from the simulation results. GSA was then applied as a screening tool to identify the input parameters having the most significant impact on each target. The results were further assessed by investigating the impact of individual parameter variations on the targets. Overall, it was demonstrated that GSA can be an effective tool in understanding parameters sensitive to a low temperature combustion concept with novel fuels.

Pal, Pinaki↗

Detailed Chemical Kinetics Based Simulation of Detonation-Containing Flows

Reliable and robust simulations of detonations in inhomogeneous and turbulent environments are of direct importance in the design of rotating detonation engines (RDEs). In particular, computational models will be especially useful in designing and optimizing discrete injectors that introduce fuel and air separately into the detonation chamber, but ensure appropriate level of mixing to sustain detonations but minimize backflow of detonation products and pressure waves into the feed plenums. Since the structure of detonations itself is non-ideal, models have to include a detailed description of this reacting flow in order to be predictive in nature. Here, a highly-scalable open source based solver has been developed for complex detonating flows such that a) the detonation processes are described using detailed chemical kinetics, b) the method is computationally efficient through the introduction of adaptive mesh refinement, and c) the solver can handle complex geometries of relevance to RDE design. Grid convergence of key metrics for detonations is evaluated using canonical flows. Further, the importance of the use of detailed chemical kinetics is illustrated by extracting the composition structure behind a two-dimensional detonation front. Finally, simulations of a practical RDE configuration are used to demonstrate the applicability of this solver to analyzing geometries. The simulation captures the general trends of the experiment well. It is found that the detonation occurs under partially-premixed conditions. Propagation of pressure waves to the injection system is observed which could influence flow behavior in the oxidizer plenum.

03 NATURAL GAS↗

Numerical Simulation of a Natural Convection–Driven Air-Cooled Reactor Cavity Cooling System Experiment

Ensuring the efficient removal of decay heat from the reactor vessel is essential for the safety of advanced reactor technologies. Several Generation-IV concepts incorporate variations in the reactor vessel cooling systems to achieve this objective. High-temperature gas-cooled reactors utilize a reactor cavity cooling system (RCCS), a passive ex-vessel system designed to operate without active components or external power during accident conditions. The RCCS removes decay heat primarily through radiative and convective heat transfer mechanisms. Here, this study presents a comprehensive validation of a computational fluid dynamics Reynolds-averaged Navier-Stokes model for the University of Wisconsin-Madison air-cooled RCCS facility. Validation was conducted for both high- and low-power natural convection cases under a uniform heating profile. Near-wall resolution was found to be critical for accurately modeling natural convection in the RCCS; employing an all-𝑦 + wall treatment resulted in wall temperature discrepancies exceeding 50 °⁢𝐶 compared to a wall-resolved mesh. Thermal-hydraulic behaviors under natural and forced convection conditions were compared within the heated cavity and RCCS. A turbulence model sensitivity analysis indicated that low-Reynolds number k-ɛ, k-ω shear stress transport (SST), and Reynolds stress transport models produce similar wall temperature predictions. A buoyancy modeling sensitivity study revealed that the Boussinesq approximation significantly underpredicted thermal-hydraulic behavior in the RCCS. Based on these findings, modeling recommendations are provided. The validated data set along with identified sensitivities refine the modeling of natural convection in the RCCS. The information produced by this study supports RCCS design, optimization, and safety evaluations, enabling the calibration and verification of reduced-order thermal-hydraulic models.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

Embedded training of neural-network subgrid-scale turbulence models

We report that the weights of a deep neural-network model are optimized in conjunction with the governing flow equations to provide a model for subgrid-scale stresses in a temporally developing plane turbulent jet at Reynolds number Re 0 = 6000 . The objective function for training is first based on the instantaneous filtered velocity fields from a corresponding direct numerical simulation, and the training is by a stochastic gradient descent method, which uses the adjoint Navier-Stokes equations to provide the end-to-end sensitivities of the model weights to the velocity fields. In-sample and out-of-sample testing on multiple dual-jet configurations show that its required mesh density in each coordinate direction for prediction of mean flow, Reynolds stresses, and spectra is half that needed by the dynamic Smagorinsky model for comparable accuracy. The same neural-network model trained directly to match filtered subgrid-scale stresses, without the constraint of being embedded within the flow equations during the training, fails to provide a qualitatively correct prediction. The coupled formulation is generalized to train based only on mean-flow and Reynolds stresses, which are more readily available in experiments. The mean-flow training provides a robust model, which is important, though a somewhat less accurate prediction for the same coarse meshes, as might be anticipated due to the reduced information available for training in this case. The anticipated advantage of the formulation is that the inclusion of resolved physics in the training increases its capacity to extrapolate. This is assessed for the case of passive scalar transport, for which it outperforms established models due to improved mixing predictions.

42 ENGINEERING↗

Finite-element modelling of multilayer X-ray optics

Multilayer optical elements for hard X-rays are an attractive alternative to crystals whenever high photon flux and moderate energy resolution are required. Prediction of the temperature, strain and stress distribution in the multilayer optics is essential in designing the cooling scheme and optimizing geometrical parameters for multilayer optics. The finite-element analysis (FEA) model of the multilayer optics is a well established tool for doing so. Multilayers used in X-ray optics typically consist of hundreds of periods of two types of materials. The thickness of one period is a few nanometers. Most multilayers are coated on silicon substrates of typical size 60 mm × 60 mm × 100–300 mm. The high aspect ratio between the size of the optics and the thickness of the multilayer (10 7 ) can lead to a huge number of elements for the finite-element model. For instance, meshing by the size of the layers will require more than 10 16 elements, which is an impossible task for present-day computers. Conversely, meshing by the size of the substrate will produce a too high element shape ratio (element geometry width/height > 10 6 ), which causes low solution accuracy; and the number of elements is still very large (10 6 ). In this work, by use of ANSYS layer-functioned elements, a thermal-structural FEA model has been implemented for multilayer X-ray optics. The possible number of layers that can be computed by presently available computers is increased considerably.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

SU-D-19A-07: Dosimetric Comparison of HDR Plesiotherapy and Electron Beam Therapy for Superficial Lesions

Purpose: Large superficial (skin, soft tissue sarcoma) lesions located on curved areas are hard to treat with electrons. The Freiburg Flap (Nucletron, Netherlands) is a flexible mesh style surface which can be easily shaped to fit curved surfaces for reproducible HDR fraction delivery. To understand the fundamental dosimetric differences, a dosimetric comparison was made between HDR plesiotherapy (Freiburg applicator for lesions over 4cm) and external electron beam radiotherapy over cases with varying target curvature (both stylized and clinical cases). Methods: Four stylized cases with variable complexity were created using artificial DICOM axial CT slices and RT structures (a square and three curved structures on a 4.5cm radius cylinder). They were planned using Oncentra v4.3 and exported to Pinnacle v9.6 for electrons planning. The HDR source dwell positions were optimized for the best coverage of the targets using graphical optimization. Electron treatment plans were created in Pinnacle using the same CT and RT structures of three HDR cases with surface lesions previously treated with the Freiburg flap. The En face electron plans used 6-12 MeV electrons and 0.5–1 cm bolus was added to increase surface dose. The electron plans were prescribed to an isodose line to conform to the target. Results: For all lesions, the average target dose coverage was similar (D90ave of 100% for HDR vs 101% for electrons). For lesions with high curvature, the HDR coverage was better (D90 102% vs D90 97% for electron). For all cases, adjacent structures high dose region was lower for HDR than electrons (D1cc 100% for HDR vs D1cc 111% for electrons). Conclusion: HDR plesiotherapy offers excellent target conformity for superficial targets similar to electrons. However, for lesions with complex curved surfaces, HDR has the advantage to achieve better dose distributions using graphical optimization to spare adjacent normal tissue while maximizing target coverage.

60 APPLIED LIFE SCIENCES↗

User-Oriented Improvements in the MOOSE framework in support of Multiphysics Simulation

The MOOSE Framework is a foundational capability used by NEAMS to create over 15 different simulation tools for advanced nuclear reactors. Due to this ubiquity, improvements to the framework in support of modeling and simulation goals are critical to the program. These improvements can take many forms including optimization, improved user experience, streamlined APIs, parallelism, and new capability. The work transcribed in the report was in direct support of the simulation tools and is already deployed or will be deployed in the coming months. The capabilities implemented were, in the same order as this report, chainable execution objects or executors, support for transfers between applications at the same level in a coupling scheme, support for boundary/subdomain restricted transfers, support for transfers between applications with different coordinate or unit systems, support for MOOSE applications in the NEAMS workbench, deployment of MOOSE application of the INL HPC OnDemand platform, addition of a triangular meshing library in libMesh and increased support of face variables.

97 MATHEMATICS AND COMPUTING↗

Flexible Computer Science Infrastructure (FleCSI)

FleCSI is a compile-time configurable framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. Current support includes multi-dimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures. FleCSI also introduces a functional programming model with control, execution, and data abstractions that are consistent with both MPI and state-of-the-art task-based runtimes such as Legion and Charm++. The FleCSI abstraction layer provides the developer with insulation from the underlying runtime, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. The control and execution models in FleCSI also provide formal nomenclature for describing poorly understood concepts like kernels and tasks.

Moss, Nicholas↗

Flexible Computer Science Infrastructure (FleCSI)

FleCSI is a compile-time configurable framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. Current support includes multi-dimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures. FleCSI also introduces a functional programming model with control, execution, and data abstractions that are consistent with both MPI and state-of-the-art task-based runtimes such as Legion and Charm++. The FleCSI abstraction layer provides the developer with insulation from the underlying runtime, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. The control and execution models in FleCSI also provide formal nomenclature for describing poorly understood concepts like kernels and tasks.

Bergen, Ben↗

Progressive collapse analysis of stiffened panel structure with consideration of actual welding distortion and residual stresses

To date, inclusion of welding distortion and residual stresses in determining the strength of marine structures is based on empirical formulations due to welding complexity. In this study, a more feasible numerical model was developed from the perspective of inherent strain theory. The accuracy of this simplified numerical model was confirmed by transient thermo-elastic–plastic finite element analysis using a novel dynamic mesh refining and iterative substructure method which is 50 times faster than conventional finite element method. Welding distortion by the two numerical approaches was validated with experimental measurement. The collapse analyses indicated that the inherent strain model predicted a compressive strength and load-displacement curve with high accuracy, as validated by the transient thermo-mechanical model. The global stiffness and the ultimate strength of the welded structure would be overestimated by about 22% if no welding effects were considered. The proposed computation procedure for determining the ultimate strength of welded structures is accurate, simple, and practical and enables improvement of ultimate strength through optimization of welding joint design to reduce the welding residual stress and distortion.

Huang, Hui↗

Structure- and Physics-Preserving Reductions of Power Grid Models

The large size of multiscale, distribution and transmission, power grids hinders fast systemwide estimation and real-time control and optimization of operations. This paper studies graph reduction methods of power grids that are favorable for fast simulations and follow-up applications. While the classical Kron reduction has been successful in reduced order modeling of power grids with traditional, hierarchical design, the selection of reference nodes for the reduced model in a multiscale, distribution and transmission, network becomes ambiguous. In this work we extend the use of the iterative Kron reduction by utilizing the electric grid's graph topology for the selection of reference nodes, consistent with the design features of multiscale networks. Additionally, we propose further reductions by aggregation of coherent subnetworks of triangular meshes, based on the graph topology and network characteristics, in order to preserve currents and build another power-flow equivalent network. Our reductions are achieved through the use of iterative aggregation of subgraphs that include general tree structures, lines, and triangles. Important features of our reduction algorithms include that (i) the reductions are either equivalent to the Kron reduction or otherwise produce a power-flow equivalent network; (ii) due to the mentioned power-flow equivalence, the reduced network can model the dynamic of the swing equations for a lossless, inductive, steady state network; (iii) the algorithms efficiently utilize hash tables to store the sequential reduction steps. The third feature allows for easy reintroduction of detailed models into the reduced, conceptual network and makes the final reduced order model backward compatible with a sequence of intermediate, partially reduced networks with varying resolution---the ordered sequence of iterative reductions corresponds to a sequence of reduced order models. The performance of our graph reduction algorithms and features of the reduced grids are discussed on a real-word transmission and distribution grid. We produce visualizations of the reduced models through open source libraries and release our reduction algorithms with example code and toy data.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Impact of LWR assembly structural features on cladding burst behavior under LOCA conditions

This provides an initial scoping study on clad balloon and burst behavior for burnup extension of reactor fuel. The associated issues with burnup extension are fuel fragmentation, relocation, and dispersal in the event of cladding failure. The general finding of this work is that the structural features, spacer grids and mixing vanes, locally suppress cladding deformation but have little impact on the overall clad performance during loss-of-coolant accidents. The work detailed in a previous report by Capps et al. focused on core optimization via neutronics, thermal hydraulics and thermomechanical analysis for burnups beyond 62 GWD/tU and enrichments above 5%. Uncertainty of fuel fragmentation relocation and dispersal in high-burnup rods during accident conditions was also investigated. The dispersal aspect of fuel fragmentation depends on cladding rupture. Thus, assessing uncertainties in the rupture behavior is helpful in estimating the dispersal of high-burnup fuel. This study builds on the previous work by assessing the impact of assembly structural features on cladding balloon and burst behavior in a full-length fuel rod. In this work, the BISON fuel performance code was used to generate 2D radial and height meshes containing structural features commonly used in nuclear fuel assemblies. First, meshes were generated with spacer grids. Results were then compared to the cladding burst temperature and balloon strain results from the previous work. A mesh sensitivity study was performed to ensure that mixing vanes and spacer grid effects were appropriately considered, resulting in a more refined mesh than the previous study. The balloon deformation and burst times of the cladding were compared to the original case. Consideration was also given to the effect of rod initial pressure. In conclusion, 3D quarter rod simulations were also performed and found good agreement with the 2D simulations in clad deformation and reasonable agreement in burst times.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

MAGNETO-FRICTIONAL MODELING OF CORONAL NONLINEAR FORCE-FREE FIELDS. I. TESTING WITH ANALYTIC SOLUTIONS

We report our implementation of the magneto-frictional method in the Message Passing Interface Adaptive Mesh Refinement Versatile Advection Code (MPI-AMRVAC). The method aims at applications where local adaptive mesh refinement (AMR) is essential to make follow-up dynamical modeling affordable. We quantify its performance in both domain-decomposed uniform grids and block-adaptive AMR computations, using all frequently employed force-free, divergence-free, and other vector comparison metrics. As test cases, we revisit the semi-analytic solution of Low and Lou in both Cartesian and spherical geometries, along with the topologically challenging Titov–Démoulin model. We compare different combinations of spatial and temporal discretizations, and find that the fourth-order central difference with a local Lax–Friedrichs dissipation term in a single-step marching scheme is an optimal combination. The initial condition is provided by the potential field, which is the potential field source surface model in spherical geometry. Various boundary conditions are adopted, ranging from fully prescribed cases where all boundaries are assigned with the semi-analytic models, to solar-like cases where only the magnetic field at the bottom is known. Our results demonstrate that all the metrics compare favorably to previous works in both Cartesian and spherical coordinates. Cases with several AMR levels perform in accordance with their effective resolutions. The magneto-frictional method in MPI-AMRVAC allows us to model a region of interest with high spatial resolution and large field of view simultaneously, as required by observation-constrained extrapolations using vector data provided with modern instruments. The applications of the magneto-frictional method to observations are shown in an accompanying paper.

79 ASTRONOMY AND ASTROPHYSICS↗

FleCSPH notes

FleCSI is a compile-time con gurable framework designed to support multi-physics application development. As such, FleCSI provides a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. FleCSI currently supports multi-dimensional mesh topology, geometry, and adjacency information, as well as n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures. FleCSI introduces a functional programming model with control, execution, and data abstractions that are consistent both with MPI and with state-of-the-art, task-based runtimes such as Legion and Charm++. The abstraction layer insulates developers from the underlying runtime, while allowing support for multiple runtime systems including conventional models like asynchronous MPI. The intent is to provide developers with a concrete set of user-friendly programming tools that can be used now, while allowing exibility in choosing runtime implementations and optimization that can be applied to future architectures and runtimes. FleCSI's control and execution models provide formal nomenclature for describing poorly understood concepts such as kernels and tasks. FleCSI's data model provides a low-buy-in approach that makes it an attractive option for many application projects, as developers are not locked into particular layouts or data structure representations. FleCSI currently provides a parallel but not distributed implementation of Binary, Quad and Oct-tree topology. This implementation is base on space lling curves domain decomposition, the Morton order. The current FleCSI version requires the implementation of a driver and a specialization driver. The role of the specialization driver is to provide the data distribution. This feature is not complete in FleCSI code and we provide it. The next step will be to incorporate it directly from FleCSPH to FleCSI as we reach a good level of performance. Then the driver represent the general execution of the resolution without worrying of the data locality and communications. As FleCSI is an On-Development code the structure may change in the future and we keep track of these changes in FleCSPH.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Early experiences on the OLCF Frontier system with AthenaPK and Parthenon–Hydro

The Oak Ridge Leadership Computing Facility (OLCF) has been preparing the nation's first exascale system, Frontier, for production and end users. Frontier is based on HPE Cray's new EX architecture and Slingshot interconnect and features 74 cabinets of optimized 3rd Gen AMD EPYC CPUs for HPC and AI and AMD Instinct 250X accelerators. As a part of this preparation, “real-world” user codes have been selected to help assess the functionality, performance, and usability of the system. This article describes early experiences using the system in collaboration with the Hamburg Observatory for two selected codes, which have since been adopted in the OLCF test harness. Experiences discussed include efforts to resolve performance variability and per-cycle slowdowns. Results are shown for a performance portable astrophysical magnetohydronamics code, AthenaPK, and a mini-application stressing the core functionality of a performance portable block-structured adaptive mesh refinement framework, Parthenon-Hydro. Here, these results show good scaling characteristics to the full system. At the largest scale, the Parthenon-Hydro miniapp reaches a total of $1.7$ $\times$ $10^{13}$ zone-cycles/s on 9216 nodes (73,728 logical GPUs) at ≈92% weak scaling parallel efficiency (starting from a single node using a second-order, finite-volume method).

97 MATHEMATICS AND COMPUTING↗

s-Step Krylov Subspace Methods as Bottom Solvers for Geometric Multigrid

Geometric multigrid solvers within adaptive mesh refinement (AMR) applications often reach a point where further coarsening of the grid becomes impractical as individual sub domain sizes approach unity. At this point the most common solution is to use a bottom solver, such as BiCGStab, to reduce the residual by a fixed factor at the coarsest level. Each iteration of BiCGStab requires multiple global reductions (MPI collectives). As the number of BiCGStab iterations required for convergence grows with problem size, and the time for each collective operation increases with machine scale, bottom solves in large-scale applications can constitute a significant fraction of the overall multigrid solve time. In this paper, we implement, evaluate, and optimize a communication-avoiding s-step formulation of BiCGStab (CABiCGStab for short) as a high-performance, distributed-memory bottom solver for geometric multigrid solvers. This is the first time s-step Krylov subspace methods have been leveraged to improve multigrid bottom solver performance. We use a synthetic benchmark for detailed analysis and integrate the best implementation into BoxLib in order to evaluate the benefit of a s-step Krylov subspace method on the multigrid solves found in the applications LMC and Nyx on up to 32,768 cores on the Cray XE6 at NERSC. Overall, we see bottom solver improvements of up to 4.2x on synthetic problems and up to 2.7x in real applications. This results in as much as a 1.5x improvement in solver performance in real applications.

97 MATHEMATICS AND COMPUTING↗

Integrated Experimental and Modeling Studies of Mineral Carbonation as a Mechanism for Permanent Carbon Sequestration in Mafic/Ultramafic Rocks

A program of laboratory experiments, modeling and fieldwork was carried out at Yale University, University of Maryland, and University of Hawai‘i, under a DOE Award (DE-FE0004375) to study mineral carbonation as a practical method of geologic carbon sequestration. Mineral carbonation, also called carbon mineralization, is the conversion of (fluid) carbon dioxide into (solid) carbonate minerals in rocks, by way of naturally occurring chemical reactions. Mafic and ultramafic rocks, such as volcanic basalt, are natural candidates for carbonation, because the magnesium and iron silicate minerals in these rocks react with brines of dissolved carbon dioxide to form carbonate minerals. By trapping carbon dioxide (CO 2 ) underground as a constituent of solid rock, carbonation of natural basalt formations would be a secure method of sequestering CO 2 captured at power plants in efforts to mitigate climate change. Geochemical laboratory experiments at Yale, carried out in a batch reactor at 200°C and 150 bar (15 MPa), studied carbonation of the olivine mineral forsterite (Mg 2 SiO 4 ) reacting with CO 2 brines in the form of sodium bicarbonate (NaHCO 3 ) solutions. The main carbonation product in these reactions is the carbonate mineral magnesite (MgCO 3 ). A series of 32 runs varied the reaction time, the reactive surface area of olivine grains and powders, the concentration of the reacting fluid, and the starting ratio of fluid to olivine mass. These experiments were the first to study the rate of olivine carbonation under passive conditions approaching equilibrium. The results show that, in a simple batch reaction, olivine carbonation is fastest during the first 24 hours and then slows significantly and even reverses. A natural measure of the extent of carbonation is a quantity called the carbonation fraction, which compares the amount of carbon removed from solution, during a run, to the maximum amount that could have been removed if the olivine initially present had fully dissolved and the cations released had subsequently precipitated in carbonate minerals. The carbonation fractions observed in batch experiments with olivine grains and powders varied significantly, from less than 0.01 (1%) to more than 0.5 (50%). Over time, the carbonation fractions reached an upper limit after about 24 to 72 hours of reaction, then stayed constant or decreased. The peak Final Scientific/Technical Report DE-FE0004275 | Mineral Carbonation | 4 coincided with the appearance of secondary magnesium-bearing silicate minerals, whose formation competes for magnesium ions in solution and can even promote conditions that dissolve magnesite. The highest carbonation fractions resulted from experiments with low ratios of concentrated solution to olivine, during which amorphous silica spheres or meshes formed, instead of secondary silicate minerals. The highest carbonation fractions appear to result from competing effects. Precipitation of silica layers on olivine reduces the reactive surface area and, thus, the rate of olivine dissolution (which ultimately limits the carbonation rate), but these same silica layers can also inhibit the formation of secondary silicate minerals that consume magnesite formed in earlier stages of carbonation. Simulation of these experiments with simple geochemical models using the software program EQ3/6 reproduces the general trends observed—especially the results for the carbonation fraction in short-run experiments. Although further experimentation and better models are needed, this study nevertheless provides a framework for understanding the optimal conditions for sequestering carbon dioxide by reacting CO 2 -bearing fluids with rocks containing olivine minerals. A series of experiments at the Rock Physics Laboratory at the University of Maryland studied the carbonation process during deformation of thermally cracked olivine-rich rock samples (dunite) saturated with CO 2 brines of varying compositions. A goal of these geomechanical experiments was to see if flow and deformation processes, which accompany natural carbonation reactions in underground settings, work to enhance or inhibit the reactions. The experiments involved hydrostatic compaction, followed by deformation at a constant rate of strain. Sample permeability was monitored during the reactions. Comparison of the samples’ volume changes to their axial strains (shortening) during deformation indicates that samples reacted with CO 2 -saturated brines accommodate more axial compaction, before the onset of dilation (a swelling that precedes rock failure), than samples reacted with distilled water. Analyses of the reacted samples with scanning electron microscope (SEM) images indicate, first, that dissolution of olivine occurring in the initial stages of carbonation can provide pathways to fluid flow that sustain the reaction, and, second, that carbonate minerals precipitated along existing fractures in the rocks may serve as asperities, or roughness on a crack’s surface that restricts its closure. Final Scientific/Technical Report DE-FE0004275 | Mineral Carbonation | 5 In a related study undertaken by one of the principal investigators as a spin-off of the main project, a simple model of (magnesite) crystal growth in the pore space of basalts undergoing carbonation was developed. The model suggests that, under a carefully controlled program of CO 2 injection, carbonate mineral growth can harden the rock formation against earthquakes that might otherwise be induced by the injection of large fluid volumes (Yarushina and Bercovici, 2013). The overall conclusion of the research project is that mineral carbonation of underground mafic and ultramafic rock formations is a viable candidate for long-term sequestration of man-made carbon dioxide. No results obtained during the project indicate that the method is inherently intractable in its implementation; moreover, enormous volumes of basalt near Earth’s surface are candidate locations for large-scale injection programs. The geochemical experiments do indicate, however, that there will be significant engineering challenges in maintaining high rates of carbonation, by delaying the onset of chemical conditions that promote formation of secondary silicate minerals and, therefore, slow down, or even reverse, the carbonation process. It remains an open question as to whether carbonation processes can be sustained for many years in an engineered system operating on a large scale—a scale capable of accommodating millions of tons of CO 2 annually. The development of realistic theoretical models that can systematically describe the combined effects of reactive flow, precipitation and geomechanical deformation is a major barrier to further understanding of the practical viability of mineral carbonation as large-scale method of carbon sequestration.

58 GEOSCIENCES↗