SEARCH · Engineering Papers
Results for “Explainability”
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.
Structure and Dynamics of Water in Polysaccharide (Alginate) Solutions and Gels Explained by the Core–Shell Model
Not Available
Machine Learning-Accelerated First-Principles Molecular Dynamics Explains Anomalous Lattice Thermal Expansion in BaZr 0.78 Y 0.22 O 3-δ
Fuel cells are a vital clean energy technology that converts chemical energy directly into electricity with high efficiency, making them a cornerstone of a sustainable energy future. Herein we investigate the thermal and chemical lattice expansion behavior of hydrated BaZr 0.78 Y 0.22 O 3-δ using machine learning-accelerated ab initio molecular dynamics simulations. Here, our results reproduce the experimentally observed non-monotonic and anomalous temperature dependence of lattice expansion, which we attribute to the competing effects of thermal expansion and dehydration—two mechanisms that influence the lattice expansion in opposite directions. The importance of this work lies in its detailed demonstration of how advanced computational techniques can accurately capture complex environmental effects, providing a valuable framework for modeling similar phenomena in a variety of material systems and applications.
Hard–Soft Acid–Base Theory Explains Photoexcited Carrier Dynamics in Porphyrin/CNT Nanohybrids: Time-Domain Atomistic Analysis
We employ the fundamental chemical concepts of hard− soft acid−base to formulate general principles governing excited-state dynamics in zinc porphyrin (ZnP)/carbon nanotube (CNT) hybrids for energy photoconversion. Atomistic quantum dynamics simulations demonstrate that electron-withdrawing and donating substituents at the ZnP β-pyrrolic position strongly influence the dynamics. ZnP photoexcitation produces subpicosecond electron transfer (ET) from ZnP to CNT, in agreement with the experiment. Substitutions of CN by H and tBu accelerate the ET. The trend is directly related to the hard−soft acid−base concept because the soft−soft interaction between the t Bu- ZnP acid and the mild CNT base enhances the donor−acceptor coupling. Longer coherence and more active vibrational modes facilitate the ET in t Bu-ZnP/CNT. Electron−hole recombination in CN-ZnP/CNT occurs on a hundred picosecond time scale, nicely corroborated by the experiment. The exciton lifetime is extended beyond a nanosecond by the substitutions. The soft−soft interaction in t Bu-ZnP/ CNT increases the splitting between the highest occupied orbitals of the two subsystems, reduces their mixing, and decreases nonadiabatic coupling between the ground and excited states. Rapid decoherence and involvement of low-frequency vibrations favor longer lifetimes. Our investigation reveals that larger p K a of the β-pyrrolic acid gives rapid ET and slow recombination and provides detailed mechanistic information, essential for future optoelectronic applications.
(Un)Explained EMIC Waves: Understanding Quiet Time EMIC Wave Drivers
Electromagnetic ion cyclotron (EMIC) waves are known to be generated through cyclotron resonance with the local ion particle population and grow when there is a large enough temperature anisotropy. In general, these temperature anisotropies necessary for wave growth are found to be associated with either solar wind pressure pulses or particle injections during geomagnetic storms or substorms. However, some EMIC events do not show any clear association with these known drivers and appear unexplained. In our analysis of high-amplitude (>1 nT) non-storm time EMIC waves, we find that 24 out of the 223 (~11%) EMIC events with peak amplitude greater than 1 nT were not found to be associated with any clear EMIC wave driver. This raises a compelling question: What magnetospheric or solar wind driver provides the free energy to grow these quiet-time EMIC waves? Here, we examine two EMIC events on 13 April 2017 and 13 February 2014, which were excited during extremely quiet solar wind and geomagnetic conditions. Furthermore, an in-depth analysis of field and particle measurements from multiple data sets, including ground and in situ data for these two events, indicates that extremely weak and otherwise insignificant pressure values and/or very weak substorm injections occurring multiple hours before the event play a significant role in quiet time wave generation.
Pivotal Role of Cloud‐Planetary Boundary Layer Coupling to Explain Contrasting Aerosol‐Cloud Relationships
The radiative effect of aerosol on cloud albedo via altering cloud droplet effective radius (r e ) is a major uncertainty in the Earth's climate system. Remote sensing studies have reported either negative or positive relationships between re and aerosol number concentration (N a ) or other aerosol proxies. However, there are much fewer in situ observational evidences and physical explanation remains elusive for the contrasting N a -r e relationships. Here we quantify the N a -r e relationship by using in situ aircraft measurements, together with a re decomposition method. Our analysis reveals that the cloud-planetary boundary layer (PBL) coupling plays a pivotal role on the N a -r e relationship. Quantitative r e decomposition indicates that the contrasting N a -r e relationships in two cloud-PBL coupling regimes result from different balances of four distinct aspects. The widely recognized number effect may be outweighed by the joint effects of the remaining three that have been rarely investigated and largely ignored in N a -r e parameterizations.
Baseflow Identification via Explainable AI With Kolmogorov‐Arnold Networks
Abstract Hydrological models often involve constitutive laws that may not be optimal in every application. We propose to replace such laws with the Kolmogorov‐Arnold networks (KANs), a class of neural networks designed to identify symbolic expressions. We demonstrate KAN's potential on the problem of baseflow identification, a notoriously challenging task plagued by significant uncertainty. KAN‐derived functional dependencies of the baseflow components on the aridity index outperform their original counterparts; they demonstrate that water availability, rather than potential evapotranspiration, drives baseflow by constraining actual evapotranspiration under arid conditions. On a test set, they increase the Nash‐Sutcliffe efficiency (NSE) by 65%, decrease the root mean squared error by 29%, and increase the Kling‐Gupta efficiency by 34%. This superior performance is achieved while reducing the number of fitting parameters from three to two. Next, we use data from 378 catchments across the continental United States to refine the water‐balance equation at the mean‐annual scale. The KAN‐derived equations based on the refined water balance outperform both the current aridity index model, with up to a 105% increase in NSE, and the KAN‐derived equations based on the original water balance. While the performance of our model and tree‐based machine learning methods is similar, KANs offer the advantage of simplicity and transparency and require no specific software or computational tools. This case study focuses on the aridity index formulation, but the approach is flexible and transferable to other hydrological processes. Plain Language Summary Equations used in hydrologic model are often suboptimal, resulting in reduced prediction accuracy and efficiency. We implemented Kolmogorov‐Arnold networks (KAN), a machine learning algorithm for deriving symbolic formulations, to estimate groundwater recharge and showed that it outperforms an existing state‐of‐the‐art semi‐empirical formulation. In hydrology, Nash‐Sutcliffe efficiency (NSE), root mean squared error (RMSE), and Kling‐Gupta efficiency (KGE) are commonly used to evaluate model performance. Higher NSE and KGE values indicate better performance, while lower RMSE values are preferable. Our results show that NSE increased by 71%, RMSE decreased by 32%, and KGE improved by 25%. In addition, KAN identifies an optimal functional form and can be used to derive new analytical formulas using the prior knowledge. The KAN‐inspired equation outperformed the original formulation and reduced the fitting parameters. Furthermore, we refined the water‐balance equation at the mean‐annual scale and showed that, based on the new water‐balance equation, KAN can derive new formulations that are superior to the original aridity index formulations (up to 105% increase in NSE) and KAN‐derived equations based on the original water balance. These findings highlight the significant potential of KAN to advance the scientific understanding of a wide range of hydrologic processes. Key Points Kolmogorov‐Arnold networks (KANs) enhance interpretability of machine‐learned hydrological models KAN‐derived symbolic formulations outperform state‐of‐the‐art semi‐empirical aridity indices KAN‐identified functional form yields an analytical index with fewer fitting parameters and improved performance
Sequence-specific dynamic DNA bending explains mitochondrial TFAM’s dual role in DNA packaging and transcription initiation
Abstract Mitochondrial transcription factor A (TFAM) employs DNA bending to package mitochondrial DNA (mtDNA) into nucleoids and recruit mitochondrial RNA polymerase (POLRMT) at specific promoter sites, light strand promoter (LSP) and heavy strand promoter (HSP). Herein, we characterize the conformational dynamics of TFAM on promoter and non-promoter sequences using single-molecule fluorescence resonance energy transfer (smFRET) and single-molecule protein-induced fluorescence enhancement (smPIFE) methods. The DNA-TFAM complexes dynamically transition between partially and fully bent DNA conformational states. The bending/unbending transition rates and bending stability are DNA sequence-dependent—LSP forms the most stable fully bent complex and the non-specific sequence the least, which correlates with the lifetimes and affinities of TFAM with these DNA sequences. By quantifying the dynamic nature of the DNA-TFAM complexes, our study provides insights into how TFAM acts as a multifunctional protein through the DNA bending states to achieve sequence specificity and fidelity in mitochondrial transcription while performing mtDNA packaging.
Aldehyde cool-flame chemistry explains a missing source of organic acids
Combustion emission is a significant source of organic acids, impacting atmospheric chemistry and climate. Their formation mechanisms, however, remain poorly understood, leading to underestimation in kinetic models. We investigate the cool-flame oxidation of key combustion intermediates—C 1 –C 4 aldehydes and benzaldehyde. Using in-situ synchrotron vacuum ultraviolet photoionization mass spectrometry, we observe the direct conversion of aldehydes to organic acids, a process enhanced by HO 2 radicals. Quantum chemistry calculations reveal that the reaction of RC(O)O 2 with HO 2 on the singlet potential energy surface contributes to organic acids. Incorporating this pathway into a kinetic model significantly improves organic acid prediction. Despite the high-temperature nature of engine combustion, significant spatial and temporal inhomogeneities (e.g., near-wall regions and crevice volumes) lead to localized cool-flame conditions, facilitating organic acid formation and emission. Elucidating the acid formation under cool-flame conditions provides a critical mechanism for accurately modelling anthropogenic organic acid emissions and developing mitigation strategies.
Can ferric-oxyl excited states explain elongated iron-oxygen bonds in heme peroxidase catalytic intermediates?
The use of X-ray structures to determine and interpret the ferryl iron-oxygen bond order in molecular oxygen-activating heme enzymes has, in the past, been controversial. This has mainly stemmed from the susceptibility of ferryl species to X-ray-induced electronic state changes. In this work we establishe using time-resolved serial femtosecond X-ray crystallography (tr-SFX) on a dye-decolourising peroxidase that the ferryl intermediate species (Compounds I and II) captured following in situ mixing of microcrystals with H 2 O 2 have single, rather than the double bond character expected. X-ray emission validated tr-SFX data with quantum refinement, time-dependent-DFT calculations and QM/MM geometry optimizations together support the concept that the single iron-oxygen bond character is not an indication of ferryl reduction or a protonated form (Fe IV -OH) but is instead attributed to the existence of accessible excited states possessing ferric-oxyl (Fe III –O •– ) character. Such states offer insight into the nature of ferryl heme.
Ion correlations explain kinetic selectivity in diffusion-limited solid-state synthesis reactions
Establishing viable solid-state synthesis pathways for novel inorganic materials remains a major challenge in materials science. Previous pathway design methods using pairwise reaction approaches have navigated the thermodynamic landscape with first-principles data but lack kinetic information, limiting their effectiveness. This gap leads to suboptimal precursor selection and predictions, especially for reactions forming competing phases with similar formation energies, where ion diffusion is a critical influence. Here we demonstrate an inorganic synthesis framework by incorporating machine learning-derived transport properties through ‘liquid-like’ product layers into a thermodynamic cellular reaction model. In the Ba–Ti–O system, known for its competitive polymorphism, we obtain accurate predictions of phase formation with varying BaO:TiO2 ratios as a function of time and temperature. We find that diffusion–thermodynamics interplay governs phase compositions, with cross-ion transport coefficients critical for predicting diffusion-limited selectivity. This work bridges length scales and timescales by integrating solid-state reaction kinetics with first-principles thermodynamics and spatial reactivity.
Emergent actin flows explain distinct modes of gliding motility
During host infection, Toxoplasma gondii and related unicellular parasites move using gliding, which differs fundamentally from other known mechanisms of eukaryotic cell motility. Gliding is thought to be powered by a thin layer of flowing filamentous (F)-actin sandwiched between the plasma membrane and a myosin-covered inner membrane complex. How this surface actin layer drives the various gliding modes observed in experiments—helical, circular, twirling and patch, pendulum or rolling—is unclear. Here we suggest that F-actin flows arise through self-organization and develop a continuum model of emergent F-actin flow within the confines provided by Toxoplasma geometry. In the presence of F-actin turnover, our model predicts the emergence of a steady-state mode in which actin transport is largely directed rearward. Removing F-actin turnover leads to actin patches that recirculate up and down the cell, which we observe experimentally for drug-stabilized actin bundles in live Toxoplasma gondii parasites. These distinct self-organized actin states can account for observed gliding modes, illustrating how different forms of gliding motility can emerge as an intrinsic consequence of the self-organizing properties of F-actin flow in a confined geometry.
Genetic variation at transcription factor binding sites largely explains phenotypic heritability in maize
Abstract Comprehensive maps of functional variation at transcription factor (TF) binding sites (cis-elements) are crucial for elucidating how genotype shapes phenotype. Here, we report the construction of a pan-cistrome of the maize leaf under well-watered and drought conditions. We quantified haplotype-specific TF footprints across a pan-genome of 25 maize hybrids and mapped over 200,000 variants, genetic, epigenetic, or both (termed binding quantitative trait loci (bQTL)), linked tocis-element occupancy. Three lines of evidence support the functional significance of bQTL: (1) coincidence with causative loci that regulate traits, includingvgt1,ZmTRE1and the MITE transposon nearZmNAC111under drought; (2) bQTL allelic bias is shared between inbred parents and matches chromatin immunoprecipitation sequencing results; and (3) partitioning genetic variation across genomic regions demonstrates that bQTL capture the majority of heritable trait variation across ~72% of 143 phenotypes. Our study provides an auspicious approach to make functionalcis-variation accessible at scale for genetic studies and targeted engineering of complex traits.
A conceptual model explaining spatial variation in soil nitrous oxide emissions in agricultural fields
Not Available
Designing green chemicals by predicting vaporization properties using explainable graph attention networks
Computational predictions of vaporization properties aid the de novo design of green chemicals, including clean alternative fuels, working fluids for efficient thermal energy recovery, and polymers that are easily degradable and recyclable.
ACES-GNN: can graph neural network learn to explain activity cliffs?
Graph Neural Networks (GNNs) have revolutionized molecular property prediction by leveraging graph-based representations, yet their opaque decision-making processes hinder broader adoption in drug discovery. This study introduces the Activity-Cliff-Explanation-Supervised GNN (ACES-GNN) framework, designed to simultaneously improve predictive accuracy and interpretability by integrating explanation supervision for activity cliffs (ACs) into GNN training. ACs, defined by structurally similar molecules with significant potency differences, pose challenges for traditional models due to their reliance on shared structural features. By aligning model attributions with chemist-friendly interpretations, the ACES-GNN framework bridges the gap between prediction and explanation. Validated across 30 pharmacological targets, ACES-GNN consistently enhances both predictive accuracy and attribution quality for ACs compared to unsupervised GNNs. Our results demonstrate a positive correlation between improved predictions and accurate explanations, offering a robust and adaptable framework to better understand and interpret ACs. This work underscores the potential of explanation-guided learning to advance interpretable artificial intelligence in molecular modeling and drug discovery.
Analytic rate theory of polariton relaxation that explains long polariton lifetime
Hybridization of a molecular exciton with a quantized photon creates a polariton. Despite extensive experimental investigations, the apparent lifetime of the exciton–polariton is not well-understood. For this work, we examined the steady-state population dynamics for a Holstein–Tavis–Cumming Hamiltonian to illuminate the long-term polaritonic dynamics and lifetime of the exciton–polariton in an optical cavity. For a realistic description of polariton relaxation, cavity loss and various exciton decay channels are included in the model. We found that in the presence of weak but finite exciton loss, the apparent lifetime of the lower polariton coincides with the out-of-cavity exciton lifetime and is independent of cavity-matter detuning. This is a simple explanation for the experimentally observed lifetimes for exciton polaritons and theoretically justifies the dark state reservoir hypothesis. Furthermore, if the upper polariton is initially populated, the system reaches the steady state very quickly, leading to single-exponential polariton relaxation. Starting from the lower polariton leads to a longer pre-steady-state time period, leading to double-exponential relaxation. Finally, we considered the effect of site orientational disorders and the exciton frequency disorderers. Under the collective limit, the effects of this disorder can be included in Fermi’s golden rule population dynamics without explicit sampling. For the exciton energy disorders, numerical calculations are needed. Our theoretical framework is applicable to interpret exciton–polariton experiments, especially related to the measured apparent lifetime of polaritons.
Osmotic and phoretic competition explains chemotaxic assembly and sorting
Microscale objects responding to chemical gradients by migrating toward or away from a preferred species is a simple yet constitutive mechanism by which transport occurs in biological organisms. Synthetic chemotaxis provides key physical descriptions of simplified systems that can be used in biological models, or in the creation of advanced responsive material systems. In this article, we provide a quantitative framework for understanding synthetic chemotaxis of microparticles which involves a competition between phoresis and osmosis. We present separate quantitative measurements of phoresis and osmosis acting on individual taxing particles, finding that phoresis follows the long-predicted v ∼ 1 / r 2 scaling while the osmotic contribution depends on the geometry and details of the system, and must be solved on a case-by-case basis. Through this, we are able to develop a more accurate picture of particle transport at the single particle level. Equipped with this approach, we go on to describe how high concentrations of particles in a symmetric chemical gradient grow close-packed hives that reach a steady-state size tunable through light intensity or particle size. Last, we demonstrate that mixed particles experiencing the same chemical gradient will selectively migrate toward or away depending on the nature of the particle surface, thereby locally sorting out a particular species. We anticipate these results will be important in describing both biological and synthetic chemotaxis in phoretic systems and should bring a wealth of studies that take advantage of competing osmotic flows to illicit unexpected dynamic active behavior.