Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Model reduction”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Electrified Operando -Freezing of Electrocatalytic CO 2 Reduction Cells for Cryogenic Electron Microscopy

The ability to freeze and stabilize reaction intermediates in their metastable states and obtain their structural and chemical information with high spatial resolution would be very powerful to unravel the fundamentals in many important materials technologies such as catalysis and batteries. Here, we develop an electrified operando-freezing methodology for the first time to preserve these metastable states under electrochemical reaction conditions for cryogenic electron microscopy (cryo-EM) imaging and spectroscopy. Using Cu catalysts for CO 2 reduction as a model system, we observe restructuring of the Cu catalyst in a CO 2 atmosphere while the same catalyst remains intact in an air atmosphere at the nanometer scale. Furthermore, we discover the existence of single valance Cu (1+) state and C-O bonding at the electrified liquid-solid interface of the operando-frozen samples, which are key reaction intermediates that traditional ex situ measurements fail to detect. Finally, this work highlights our novel technique to study the local structure and chemistry of electrified liquid-solid interfaces, which has broad impact for many electrochemical reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Pore-Scale Transport Effects in Electrochemical CO 2 Reduction on Gold via Coupled Microkinetic-Transport Modeling

A pore-resolved modeling framework is developed to quantify how pore-scale transport affects the intrinsic microkinetics of CO 2 -to-CO on Au. A DFT-informed microkinetic model is coupled self-consistently to a Generalized-Modified Poisson–Nernst–Planck (GMPNP) transport description in a single, electrolyte-filled cylindrical pore, allowing local concentrations and potential to feed back into site-specific reaction rates. FIB-SEM is used to determine pore sizes within realistic electrode materials. Across pore diameters, d p = 10–6000 nm, the surface-averaged CO 2 reduction rate is systematically reduced relative to the ideal microkinetic baseline where mass transport is not accounted for; the effectiveness factor 𝜂 𝑠,CO 2 , which quantifies this ratio, decreases rapidly at more negative potentials and is about 1% near −1.0 V vs SHE due to reactant depletion. Spatial maps reveal pore-bulk alkalization that emerges at higher cathodic bias, with a small, near-wall pH dip due to electrostatic repulsion of hydroxide at the cathode interface. For a fixed aspect ratio L p /d p , narrower pores exhibit larger 𝜂 𝑠,CO 2 by shortening diffusion paths, whereas variations in the aspect ratio L p /d p play a secondary role. A dimensionless analysis (surface/bulk Damköhler numbers) delineates operating regimes. In conclusion, this work offers a concept for incorporating microkinetic models into homogenized porous-electrode models through effectiveness factors and pore-size distribution.

Au-catalyst↗
DOE OSTI2025

Pseudo-viscous modeling of transport in dense granular flows for thermal energy storage applications

Dense, granular flows were examined to effectively capture and model bulk viscous properties in thin packed beds. A modified Couette cell with particle image velocimetry was used to experimentally determine pseudo-viscosity properties of four particulate media with varying morphologies: (1) iron oxide-coated SiO 2 particles, (2) CARBOBEAD CP30-60 particles, (3) CARBOBEAD CP40-100 particles, and (4) Al 2 O 3 beads. The pseudo-viscosity functions were fitted using a power law to correlate the measured shear stress as a function of measured shear rate. The pseudo-viscous functions were used as inputs to computation fluid dynamics models for a single-phase viscous fluid to predict granular flow profiles. Steady-state free surface velocity profiles at angular velocities <7 rad/s predicted by the model were in good agreement with the experimental particle image velocimetry measurements, resulting in Pearson correlation coefficients of 0.97 for iron-oxide coated SiO 2 particles and 0.95 for CP30-60 particles. As a result, this alternative approach to measuring pseudo-viscous properties under shearing and modeling bulk transport behavior of granular flow using computation fluid dynamics model offered significant reduction in computational load compared to discrete element methods.

14 SOLAR ENERGY↗
DOE OSTI2025

EPR and 31 P ENDOR Characterization of Pseudo-Jahn–Teller Dynamics and N 2 Activation in Functional Nitrogenase Models, P 3 E M(N 2 ) (M = Fe, Co; E = Si, B, C)

Here, the nominally trigonal, pseudo-Jahn-Teller (PJT)-active, S = ½ N 2 -bound transition-metal complexes, P 3 E M(N 2 ), M = Fe, Co, with three in-plane phosphine-ligands and axial donors, E = Si, B, C, include functional nitrogenase models that catalyze reduction of N 2 to NH 3 . We applied EPR, 31 P ENDOR spectroscopy and DFT computations to characterize the PJT-induced distortions of four selected P 3 E M(N 2 ), revealing how the metal-ion and axial ligand E together tune both PJT dynamics and N 2 activation for reduction. Comparisons reveal an unrecognized correlation between PJT distortion, M-E bond elasticity, and N 2 activation, providing guidelines for designing bioinspired N 2 -reduction catalysts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

p K a of alcohols dictates their reactivity with reduced uranium-substituted thiomolybdate clusters

The uranium-substituted thiomolybdate cluster, (Cp* 3 Mo 3 S 4 )UCp*, has been demonstrated as a model for water reduction by single uranium atoms supported on a molybdenum sulfide surface (U@MoS 2 ). In this study, the scope of O–H bond activation is expanded through the investigation of the reactivity of various alcohols with differing pKa values for the –OH proton. The reaction of (Cp* 3 Mo 3 S 4 )UCp* with stoichiometric amounts of methanol, phenol, 2,6-dichlorophenol, and nonafluoro-tert-butyl alcohol affords the corresponding mono-alkoxide species, (Cp* 3 Mo 3 S 4 )Cp*U(OR), via a uranium-metalloligand cooperative activation of the O–H bond. This observed reactivity is analogous to the O–H bond activation reported by (Cp* 3 Mo 3 S 4 )UCp* in the presence of water. However, addition of tert-butanol induces protonolysis of the Cp* ligand on uranium, resulting in the formation of a uranium tris-tert-butoxide cluster, (Cp* 3 Mo 3 S 4 )U(O t Bu) 3 . Independent synthesis of (Cp* 3 Mo 3 S 4 )Cp*U(O t Bu) was possible via an alternative pathway, eliminating sterics as a justification for the observed discrepancy in reactivity. Furthermore, these results offer insight into the role the –OH proton pKa plays in dictating the mechanism of O–H bond activation of alcohols by the uranium-substituted thiomolybdate cluster.

Patra, Kamaless↗
DOE OSTI2025

Measurements of radial neutral density profiles from Balmer- α emission in Wendelstein 7-X

Radial neutral density profiles are estimated from measurements of passive H α emission in the Wendelstein 7-X stellarator. To parametrize the generally three-dimensional distribution with a low number of degrees of freedom, the neutral density is reduced to a flux surface quantity. Accounting for emission from excitation and recombination processes, neutral density profiles are derived independently for each of the available lines of sight. Density profiles obtained from the different viewing geometries are found to vary within one order of magnitude. Toroidally oriented lines of sight predict systematically lower neutral densities when compared to poloidally oriented ones. This discrepancy is attributed to the simplifications inherent in the imposed model and significant differences in integration volumes across the viewing geometries. In line with expectations, obtained neutral densities are found to decrease with increasing plasma density. Key restrictions of the model include the reduction of the neutral density to a flux surface quantity, uncertainties in the plasma profiles and instrument function, and line integration effects outside the last closed flux surface.

Wendelstein 7-X↗
DOE OSTI2026

Progressing Analysis of Variable Electric Rates (PAVER) Study

The Progressing Analysis of Variable Electric Rates (PAVER) study analyzed the impact of a range of time-varying electric rates on the performance of a regional electric grid and the resulting costs for participating and non-participating customers. This analysis leveraged and extended the work of PNNL’s Distribution System Operator with Transactive (DSO+T) study. Five different rate designs were included: a flat volumetric energy charge, a typical Time of Use (TOU) rate, a dynamic energy (DE) rate (based on wholesale locational marginal prices), a dynamic energy and capacity (DE+C) rate, and, finally, a Block and Swing (B&S) rate that billed customers based on their average load profile at constant pricing, but used the DE+C dynamic price for load deviations from their average profile. These rates were analyzed in a large-scale co-simulation of an entire regional grid with a customer population representative of the current state. A large fraction (80%) of residential and commercial customers were assumed to participate in these time-varying rates with automatically controlled HVAC, water heaters, electric vehicles, and batteries. This study assumed no industrial sector participation. The DE and DE+C rates saw system peak loads reduced by 6-7%, while the large participation in the TOU rate case saw a significant rebound effect and a resulting peak load increase of >5%. The impacts to the annual and peak system demand impacted system wholesale prices and the overall grid operating costs. This cost structure determined the revenue needed to be collected from customers by each rate design. Participating customers on the DE and DE+C rates (located in one of the modeled DSOs) saw reductions in average annual electricity bills of 11-17% with average increases in monthly bill variation of no more than 13%. At such high participation levels, TOU customers saw 10% higher average annual bills (due to system-wide rebound effects) and average increased monthly bill variation of 16%. Residential owners of large flexible loads (such as electric vehicles) saw larger bill savings (17-20%) when on a fully dynamic rate. The presence of on-site generation (such as rooftop solar) did not appear to appreciably change customer outcomes. Customers on the Block and Swing rate did see 6% lower monthly bill variation (as intended) than the flat rate case, but at the expense of appreciable bill savings, which were only 3%, comparable to the savings seen by non-participants. Given this finding we recommend that additional research be conducted into how best various bill protection mechanisms can balance minimizing customer bill variation with providing financial incentives commensurate with the flexibility customers provide. We also recommend that customer outcomes be explored across a range of regions using current actual customer and system cost data.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2025

Brillouin Sensing with PCA, and PCA-Based Neural Networks for Efficient Temperature Monitoring

This work explores peak estimation techniques in Brillouin Optical Time Domain Analysis (BOTDA), emphasizing both accuracy and efficiency. Euclidean distance measurement method is applied to principal components derived from Brillouin Gain Spectrum data. It offers a major speed advantage being 180 170 times faster than traditional curve fitting methods such as Lorentzian curve fitting, while maintaining similar accuracy. Additionally, a PCA- based neural network model shows significant reduction of peak estimation time compared to Lorentzian fitting. Results show Brillouin frequency shift errors lie under 0.75 MHz in both Euclidean distance-based and neural network-based methods, both of which utilize PCA components. For large data sets and long length fibers, PCA- assisted neural network for peak estimation would be an efficient solution.

Distributed optical fiber sensing↗
DOE OSTI2026

A Generalized Grain-Scale Model for the Non-Plasma and Plasma-Assisted Hydrogen Direct Reduction of Iron Ore

Direct Reduction of Iron ore using hydrogen (H-DRI) is a promising pathway towards efficient steelmaking and accurate predictive models are a necessity for scale-up and optimization of this technology. However, accurate models of this process remain limited because existing models oversimplify grain-scale phenomena, such as nonlinearity inside grain, self-sufficient porosity, surface reactions, and the role of plasma species. These phenomena are important for flash steelmaking and plasma-assisted H-DRI processes. To address this need, we present a phenomenological model for simulating H-DRI at the scale of a single micron-sized grain of the iron ore. We call this the Transient Reactive Grain Model (TRGM). TRGM incorporates key physical process: gas species transport, a chemical kinetics of material conversion, nanopore structural evolution and, adsorption-desorption surface kinetics at the reactive nanopore surface. The important contribution of this work is that the model provides a dependence on different reductant species, specifically hydrogen atoms versus molecules, so that role of hydrogen plasma reduction can be clarified compared to the use of pure hydrogen gas reduction. TRGM predictions agree well with experimental data for both molecular H2 reduction of Fe2O3 and plasma hydrogen reduction of Fe3O4. Results reveal species concentration gradients with a diffuse reaction zone, and enhanced hydrogen diffusion at the grain outer surface due to evolving porosity. These findings challenge common assumptions in existing models, including sharp reaction fronts, quasi-steady diffusion and kinetics, and the neglect of surface chemistry. As a generalized grain-scale model for H-DRI processes, TRGM has practical applications in flash steelmaking and in-flight reduction using both molecular and plasma hydrogen.

08 HYDROGEN↗
DOE OSTI2025

Synthesis, Properties, and Electrochemical Proton Reduction of a Homoleptic Tetrathiolato Ni-Site Model of [NiFe]-Hydrogenase

[NiFe]-hydrogenase enzymes process H 2 at a nonplanar tetracysteinato-Ni site, the sole participator in proton binding/redox chemistry during turnover. With the objective of assessing whether a simple tetrahedral/ tetrathiolato-Ni 2+ core could promote H 2 evolution reaction (HER), we synthesized (Et 4 N) 2 [Ni(S-p-CF 3 −Ph) 4 ] (1) employing para-trifluoromethylbenzenethiolate ( − S-p-CF 3 −Ph) as a Ni-site analog of [NiFe]-hydrogenase. Spectroscopic measurements and X-ray crystallography confirm the distorted tetrahedral geometry of 1. Dissolution of 1 results in partial thiolate dissociation and formation of S,S-bridged complexes such as (Et 4 N) 2 [Ni 2 (S-p-CF 3 −Ph) 6 ] (3) among other ill-defined species. Dissociation is further accelerated in the presence of Brønsted acids, complicating the assessment of 1 for proton reduction. However, this dissociation/proton instability is suppressed in the presence of additional thiolate ligand to ensure tetrahedral/tetrathiolato 1 persists in solution. Electrochemical HER activity was evaluated by monitoring the current response of an MeCN solution of 1/ excess thiolate after sequential titration with a weak Brønsted acid (acetic acid). The results suggest that 1 is a modest electrocatalyst for the HER with a turnover frequency of 14.5 ± 3.6 s −1 and an overpotential of 0.72 ± 0.02 V. Control experiments and supplementary DFT computations indicate that 1, or a species derived from 1, is responsible for the HER and suggest an ECCE-type mechanism.

Evolution reactions↗
DOE OSTI2024

Non-Electricity Based Renewable Fuels: Theory and Computation for Solar Thermochemical Hydrogen

Dominated by photovoltaics and wind, current renewable energy sources generate mostly electricity, but 80% of the global final energy consumption occurs in form of fuels. Therefore, direct solar fuel generation would be a major breakthrough for the energy transition. Solar thermochemical hydrogen (STCH) is one of the very few potential routes towards scalable renewable fuels, but currently suffers from lack of an oxide working material that could optimally perform energy conversion within the thermodynamic boundary conditions. Theory and computation can contribute in two distinct ways, through materials search and discovery, but also by providing detailed mechanistic models for specific systems so to advance our understanding of possible design strategies. To enable high-throughput materials screening, we developed a defect graph neural network (dGNN) machine learning approach,[1] which accelerates the prediction of defect formation energies by replacing the tedious density functional theory (DFT) supercell calculations for all possible defect sites. This approach enables high-throughput database screening of oxides, which was integrated with thermodynamic modeling to extract the reduction entropies as additional selection criterion for STCH. Once potential candidate materials are identified, detailed models can guide materials design by predicting performance characteristics. One challenge is to quantitatively predict thermochemical equilibria at high concentrations when the redox active defects start to interact with each other, thereby impeding the formation of additional defects. Introducing a model for the free energy of defect interaction, parametrized on the basis of DFT data, we simulated the complete STCH redox cycle for (Sr,Ce)MnO3 alloys, achieving near-quantitative agreement with experimental data.[2] The analysis of these simulations reveals how defect interactions diminish the reduction entropy and H2 yield, suggesting to include these interactions in design considerations. Finally, we revisit the popular van't Hoff method for analyzing reduction enthalpies and entropies. This method is not ideal, as it involves a temperature-dependent convolution of gas-phase and solid-state entropies, causing uncertainties in the same order of magnitude as the physical quantities of interest. To avoid this problem, we suggest a simple alternative approach which can be applied to experimental and simulated data alike.

first-principles calculations↗
DOE OSTI2026

Improved Regional Moment Tensor Inversion for Moderately Large Earthquakes in the Western United States Using a 3D Earth Model Based on Full Waveform Tomography

The nature of seismic sources for moderately large (moment magnitude, M w 5.0–6.5) events are commonly characterized by their moment tensor (MT) solutions and obtained by inversion of regional distance (200–1600 km) long‐period (20–50 s) waveforms. Regional MT estimates are often calculated from average plane‐layered, one‐dimensional (1D) velocity models. However, 1D model calculations can produce misfits in the arrival times and waveform shapes that introduce errors, particularly at longer distances or for shorter periods, which are necessary for analyzing lower magnitude events. Approximate Earth models (e.g., 1D) representing broad areas may be inadequate, particularly in the crust and uppermost mantle of tectonically complex regions. In this study, we show how a three‐dimensional (3D) Earth model obtained from full waveform inversion tomography can improve waveform fits and decrease phase errors. We developed a platform and workflow to perform routine 3D MT inversions and inverted MTs for 25 earthquakes in the western United States and seven nuclear explosions using an average 1D and a recent 3D Earth model, WUS256 (Rodgers et al., 2022). Using the 3D model improves waveform fits (variance reduction and phase time shifts) compared with the 1D model, and the 3D MT solutions are stable across large distances. This study shows that 3D models obtained from full waveform tomography can improve MTs and source characterization especially at far regional distances (>800 km).

Geosciences↗
DOE OSTI2025

The DREAM approach to demand-side emissions reductions in Indonesia, 2020–2060

Indonesia’s current energy system modeling is heavily focused on the supply side, but emissions reductions in the demand sector have a significant impact on advancing Indonesia’s ambitious emissions reductions goals. To address the gap, we develop a new modeling tool DREAM Indonesia based on a bottom-up, technology-rich demand side framework and formulate projections of demand-side emissions reductions in Indonesia in 2020–2060. We find that demand-side energy efficiency and electrification can halve the growth rate of final energy demand to 1.4% annually over 2020–2060 and reverse the growing trend of emissions. The feasibility of full electrification by 2060, coupled with rapid adoption of existing technologies, positions the building sector as a model for achievable decarbonization and a cornerstone of Indonesia’s emissions reductions ambitions. In the industrial sector, extensive emissions reductions of 87% by 2060 (compared to business-as-usual) are achievable through energy efficiency improvements, alongside enhanced material efficiency measures including optimized material usage, low-carbon substitutions, innovative technologies, and increased circularity. In the transportation sector, balancing final energy demand by incorporating energy efficiency improvements across all transport modes, along with electrification particularly in road transportation, could decrease the emissions by up to 82% in 2060 compared to business-as-usual. This study provides insights and modeling approaches for rapidly growing Asian economies as well as other developing countries facing combined development and decarbonization challenges.

DREAM Indonesia↗
DOE OSTI2024

Mixing and dilution controls on marine CO 2 removal using alkalinity enhancement

Marine CO 2 removal (CDR) using enhanced-alkalinity seawater discharge was simulated in the estuarine waters of the Salish Sea, Washington, US. The high-alkalinity seawater would be generated using bipolar membrane electrodialysis technology to remove acid and the alkaline stream returned to the sea. Response of the receiving waters was evaluated using a shoreline resolving hydrodynamic model with biogeochemistry, and carbonate chemistry. Two sites, and two deployment scales, each with enhanced TA of 2997 mmol m -3 and a pH of 9 were simulated. The effects on air-sea CO 2 flux and pH in the near-field as well as over the larger estuary wide domain were assessed. The large-scale deployment (addition of 164 Mmoles TA yr -1 ) in a small embayment (Sequim Bay, 12.5 km 2 ) resulted in removal of 2066 T of CO 2 (45% of total simulated) at rate of 3756 mmol m -2 yr -1 , higher than the 63 mmol m -2 yr -1 required globally to remove 1.0 GT CO 2 yr -1 . It also reduced acidity in the bay, ΔpH ≈ +0.1 pH units, an amount comparable to the historic impacts of anthropogenic acidification in the Salish Sea. The mixing and dilution of added TA with distance from the source results in reduced CDR rates such that comparable amount 2176 T CO 2 yr -1 was removed over >1000 fold larger area of the rest of the model domain. There is the potential for more removal occurring beyond the region modeled. The CDR from reduction of outgassing between October and May accounts for as much as 90% of total CDR simulated. Of the total, only 375 T CO 2 yr -1 (8%) was from the open shelf portion of the model domain. With shallow depths limiting vertical mixing, nearshore estuarine waters may provide a more rapid removal of CO 2 using alkalinity enhancement relative to deeper oceanic sites.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Emulator-based Bayesian calibration of a subglacial drainage model

Subglacial drainage models, often motivated by the relationship between hydrology and ice flow, sensitively depend on numerous unconstrained parameters. We explore using borehole water-pressure time series to calibrate the uncertain parameters of a popular subglacial drainage model, taking a Bayesian perspective to quantify the uncertainty in parameter estimates and in the calibrated model predictions. To reduce the computation time associated with Markov Chain Monte Carlo sampling, we construct a fast Gaussian process emulator to stand in for the subglacial drainage model. We first carry out a calibration experiment using synthetic observations consisting of model simulations with hidden parameter values as a demonstration of the method. Using real borehole water pressures measured in western Greenland, we find meaningful constraints on four of the eight model parameters and a factor-of-three reduction in uncertainty of the calibrated model predictions. These experiments illustrate Gaussian process-based Bayesian inference as a useful tool for calibration and uncertainty quantification of complex glaciological models using field data. However, significant differences between the calibrated model and the borehole data suggest that structural limitations of the model, rather than poorly constrained parameters or computational cost, remain the most important constraint on subglacial drainage modelling.

58 GEOSCIENCES↗
DOE OSTI2025

Hybrid PDES Simulation of HPC Networks Using Zombie Packets

Although high-fidelity network simulations have proven to be reliable and cost-effective tools to peer into architectural questions for high-performance computing (HPC) networks, they incur a high resource cost. The time spent in simulating a single millisecond of network traffic in the highest detail can take hours, even for static, well-behaved traffic patterns such as uniform random. Surrogate models offer a significant reduction in runtime, yet they cannot serve as complete replacements and should only be used when appropriate. Thus, there is a need for hybrid modeling, where high-fidelity simulation and surrogates run side-by-side. Here, we present a surrogate model for HPC networks in which: packets bypass the network, while the network state is left untouched, i.e., suspended. To bypass the network, we use historical data to estimate the arrival time at which every packet should be scheduled at; to suspend the network, all in-flight packets are scheduled to arrive at their destinations, and are kept in the system to awaken as zombies when switching back to high-fidelity. Speedup for a hybrid model is relative to the proportion of surrogate to high-fidelity. This light-weight surrogate obtained up to 76× speedup. Keeping the zombies in the network showed an increase in the accuracy of the high-fidelity simulation on restart when compared to restarting the network from an empty state.

HPC networks↗
DOE OSTI2026

Systematic Benchmarking of Climate Models: Methodologies, Applications, and New Directions

As climate models become increasingly complex, there is a growing need to comprehensively and systematically assess model performance with respect to observations. Given the increasing number and diversity of climate model simulations in use, the community has moved beyond simple model intercomparison and toward developing methods capable of benchmarking a large number of simulations against a suite of climate metrics. Here, we present a detailed review of evaluation and benchmarking methods and approaches developed in the last decade, focusing primarily on scientific implications for Coupled Model Intercomparison Project (CMIP) simulations and CMIP6 results that contributed to the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report (AR6). Based on this review, we explain the resulting contemporary philosophy of model benchmarking, and provide clear distinctions and definitions of the terms model verification, process validation, evaluation, and benchmarking. While significant progress has been made in model development based on systematic evaluation and benchmarking efforts, some climate system biases still remain. The development of open‐source community software packages has played a fundamental role in identifying areas of significant model improvement and bias reduction. We review the key features of several software packages that have been commonly used over the past decade to evaluate and benchmark global and regional climate models. Additionally, we discuss best practices for the selection of evaluation and benchmarking metrics and for interpreting the obtained results, the importance of selecting suitable sources of reference data and accurate uncertainty quantification.

Environmental sciences↗
DOE OSTI2026

Revealing EDL-driven reduction mechanisms in binary, ternary, and quaternary fluorinated electrolytes via an integrated MD–DFT–ML framework

Accurately predicting solid electrolyte interphase (SEI) formation requires explicitly resolving the electric double layer (EDL) structure, which deviates significantly from that of the bulk electrolyte. Although an established molecular dynamics (MD) and Density Functional Theory (DFT) framework can model SEI formation by evaluating reduction reactions of local clusters in the EDL, it suffers from a combinatorial computational bottleneck. To overcome this limitation, we introduce a machine-learning-accelerated simulation workflow (MD–DFT–ML), integrating a gradient-boosted regression model trained on EDL composition data to efficiently predict reduction potentials. We apply this framework to seven fluorinated electrolytes comprising fluorinated anions, a fluorinated ester solvent, two types of diluent (ion-solvating ester vs. non-solvating ether), and an FEC additive. The analysis shows that the EDL selectively accumulates cation-binding species; consequently, the non–cation-binding ether diluent rarely enters the EDL and makes minimal contributions to SEI formation. DFT calculations on statistically representative EDL clusters provide reduction potentials and fluorine-release pathways, while the ML model, which substantially reduces the DFT workload, predicts cluster reduction energies with a mean absolute error of 0.1 eV. The combined MD–DFT–ML approach also quantifies contributions from different sources to LiF formation in the SEI. This methodology establishes a generalizable route for multiscale modeling electrolyte and interphase design for next-generation electrochemical energy-storage systems.

DFT-MD-ML workflow↗