SEARCH · Engineering Papers
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.
Embedding matrix completion methods from machine learning in classical thermodynamic models creates powerful hybrid models for predicting properties of mixtures.
The acetonyl radical (•CH 2 COCH 3 ) is relevant to atmospheric and combustion chemistry due to its prevalence in many important reaction mechanisms. One such reaction mechanism is the decomposition of Criegee intermediates in the atmosphere that can produce acetonyl radical and OH. In order to understand the fate of the acetonyl radical in these environments and to create more accurate kinetics models, we have examined the reaction system of the acetonyl radical with O 2 using highly reliable theoretical methods. Structures were optimized using coupled cluster theory with singles, doubles, and perturbative triples [CCSD(T)] with an atomic natural orbital (ANO0) basis set. Energetics were computed to chemical accuracy using the focal point approach involving perturbative treatment of quadruple excitations [CCSDT(Q)] and basis sets as large as cc-pV5Z. The addition of O 2 to the acetonyl radical produces the acetonylperoxy radical, and multireference computations on this reaction suggest it to be barrierless. Additionally, no submerged pathways were found for the unimolecular isomerization of the acetonylperoxy radical. Besides dissociation to reactants, the lowest energy pathway available for the acetonylperoxy radical is a 1-5 H shift from the methyl group to the peroxy group through a transition state that is 3.3 kcal mol -1 higher in energy than acetonyl radical + O 2 . The ultimate products from this pathway are the enol tautomer of the acetonyl radical along with O 2 . Multiple pathways that lead to OH formation are considered; however, all of these pathways are predicted to be energetically inaccessible, except at high temperatures.
We employ polymer integral equation theory to study a simplified model of semiflexible polymerized ionic liquids (PolyILs) that interact via hard core repulsions and short range screened Coulomb interactions. The multi-scale structure in real and Fourier space of PolyILs (ions chosen to mimic Li, Na, K, Br, PF 6 , and TFSI) are determined as a function of melt density, Coulomb interaction strength, and ion size. Comparisons with a homopolymer melt, a neutral polymer–solvent-like athermal mixture, and an atomic ionic liquid are carried out to elucidate the distinct manner that ions mediate changes of polymer packing, the role of excluded volume effects, and the influence of chain connectivity, respectively. The effect of Coulomb strength depends in a rich manner on ion size and density, reflecting the interplay of steric packing, ion adsorption, and charge layering. Ion-mediated bridging of monomers is found, which intensifies for larger ions. Intermediate range charge layering correlations are characterized by a many-body screening length that grows with PolyIL density, cooling, and Coulomb strength, in disagreement with Debye–Hückel theory, but in accord with experiments. Qualitative differences in the collective structure, including an ion-size-dependent bifurcation of the polymer structure factor peak and pair correlation function, are predicted. The monomer cage order parameter increases significantly, but its collective ion counterpart decreases, as ions become smaller. Such behaviors allow one to categorize PolyILs into two broad classes of small and large ions. Furthermore, dynamical implications of the predicted structural results are qualitatively discussed.
Aperiodic formations continue to focus interest in areas ranging from advanced scientific theories to practical everyday applications. Starting from diverse and tightly bonded intermetallic compounds, this world showed an important breakthrough toward the so-called soft systems of meso/macroscale: liquid crystals, thin films, polymers, proteins, etc. This work opens a route for making bulk quasicrystals (QC) in an unprecedented but very common area, with molecular ligands. Since these systems are, to a large extent, transparent, they extend the possible areas of QC application to previously unreachable corners, e.g., photonics. We combined efficient bridging ligands with uranyl pentagonal bonding centers and, unexpectedly, brought the unique attributes of f-element coordination chemistry to an interdisciplinary area of aperiodic formations. Taking advantage of the planar coordination of uranyl ions, we were able to direct the structure expansion solely in two directions with a characteristic snub square tiling, a predicted but previously unobtainable dodecagonal approximant.
The surfaces of most metals immersed in aqueous electrolytes have a several-nanometer-thick oxide/hydroxide surface layer. This gives rise to the existence of both metal|oxide and oxide|liquid electrotlyte interfaces, and makes it challenging to correlate atomic length-scale structures with electrochemical properties such the potential-of-zero-charge (PZC). The PZC has been shown to be correlated the onset potential for pitting corrosion. Here, we conduct large-scale Density Functional Theory and ab initio molecular dynamics to calculate the PZC of a Al(111)| γ -Al 2 O 3 (110)| water double-interface model within the context of aluminum corrosion. By partitioning the multiple interfaces involved into binary components with additive contributions to the overall work function and voltage, we predict the PZC to be -1.53 V vs SHE for this model. Furthermore, we calculate the orbital energy levels of defects like oxygen vacancies in the oxide, which are critical parameters in theories associated with pitting corrosion. We predict that the Fermi level at the PZC lies above the impurity defect levels of the oxygen vacancies, which are therefore uncharged at the PZC. From the PZC estimate, we predict the voltage needed to create oxygen vacancies with net positive charges within a flatband approximation.
This development of empirical data to support realistic and science-based input to safety regulations and transportation standards is a critical need for the hazardous material (HM) transportation industry. Current regulations and standards are based on the TNT equivalency model. However, real world experience indicates that use of the TNT equivalency model to predict composite overwrapped pressure vessel (COPV) potential energy release is unrealistically conservative. The purpose of this report is to characterize and quantify rupture events involving damaged COPV’s of the type used in HM transportation regulated by the Department of Transportation (DOT). This was accomplished using a series of five tests; 2 COPV tests for compressed natural gas (CNG), 2 COPV tests for hydrogen, and 1 COPV test for nitrogen. Measured overpressures from these tests were compared to predicted overpressures from a TNT equivalence model and blast curves. Comparison between the measurements and predictions shows that the predictions are generally conservative, and that the extent of conservatism is dominated by predictions of the chemical contribution to overpressure from fuel within the COPVs.
The most recent Omicron variant of SARS-CoV-2 has caused global concern and anxiety. The only thing certain about this strain, with a large number of mutations in the spike protein, is that it spreads quickly, seems to evade immune defense, and mitigates the benefits of existing vaccines. Based on the ultra-large-scale ab initio computational modeling of the receptor binding motif (RBM) and the human angiotensin-converting enzyme-2 (ACE2) interface, we provide the details of the effect of Omicron mutations at the fundamental atomic scale level. In-depth analysis anchored in the novel concept of amino acid-amino acid bond pair units (AABPU) indicates that mutations in the Omicron variant are connected with (i) significant changes in the shape and structure of AABPU components, together with (ii) significant increase in the positive partial charge, which facilitates the interaction with ACE2. We have identified changes in bonding due to mutations in the RBM. The calculated bond order, based on AABPU, reveals that the Omicron mutations increase the binding strength of RBM to ACE2. Our findings correlate with and are instrumental to explain the current observations and can contribute to the prediction of next potential new variant of concern.
The receptor-binding domain (RBD) is the essential part in the Spike-protein (S-protein) of SARS-CoV-2 virus that directly binds to the human ACE2 receptor, making it a key target for many vaccines and therapies. Therefore, any mutations at this domain could affect the efficacy of these treatments as well as the viral-cell entry mechanism. We introduce ab initio DFT-based computational study that mainly focuses on two parts: (1) Mutations effects of both Delta and Omicron variants in the RBD-SD1 domain. (2) Impact of Omicron RBD mutations on the structure and properties of the RBD-ACE2 interface system. The in-depth analysis is based on the novel concept of amino acid-amino acid bond pair units (AABPU) that reveal the differences between the Delta and/or Omicron mutations and its corresponding wild-type strain in terms of the role played by non-local amino acid interactions, their 3D shapes and sizes, as well as contribution to hydrogen bonding and partial charge distributions. Our results also show that the interaction of Omicron RBD with ACE2 significantly increased its bonding between amino acids at the interface providing information on the implications of penetration of S-protein into ACE2, and thus offering a possible explanation for its high infectivity. Our findings enable us to present, in more conspicuous atomic level detail, the effect of specific mutations that may help in predicting and/or mitigating the next variant of concern.
It has been well-established that the topography around conical intersections between excited electronic states is incorrectly described by coupled cluster and many other single reference theories (the intersections are “defective”). Despite this, we show both analytically and numerically that the geometric phase effect (GPE) is correctly reproduced upon traversing a path around a defective excited-state conical intersection (CI) in coupled cluster theory. The theoretical analysis is carried out by using a non-Hermitian generalization of the linear vibronic coupling approach. Interestingly, the approach qualitatively explains the characteristic (incorrect) shape of the defective CIs and CI seams. Moreover, the validity of the approach and the presence of the GPE indicate that defective CIs are local (and not global) artifacts. This implies that a sufficiently accurate coupled cluster method could predict nuclear dynamics, including geometric phase effects, as long as the nuclear wavepacket never gets too close to the conical intersections.
Not Available
A hydrodynamic shear mixing layer experiment at the National Ignition Facility had previously demonstrated Eulerian scaling of integrated, late-time quantities, including turbulent kinetic energy. In this manuscript, the experiment is repeated with new materials. In this work, using the new dataset, we demonstrate that Euler-number scalings hold not just for late time, but dynamically throughout the experiment, for measurements in all three spatial dimensions. Incorporating the dynamic scaling leads to an enhanced calculation that the heavier of the two scaled experiments has approached three generations of mergers of its primary instability's structures and a consistent observation of such a merger in action in the lighter of the two scaled experiments. Furthermore, the improved scrutiny of the time evolution of instability structures leads to sharper estimates of turbulent kinetic energy, including a demonstration of different behaviors correlating with surface roughness (quantitatively consistent with transitions between laminar and turbulent initial states), as predicted by a Reynolds-averaged turbulent model, which evidently correctly handles the differing shock-roughness interactions to drive its internal state of the model into different regimes. Altogether, a picture arises of the analytical improvements in treating these variations (of times, densities, and roughnesses) as a unified whole and of multiple ways by which deviations from the scaling could indicate an onset of non-hydrodynamic behavior. Such deviations were not expected for these experiments (which models correctly indicated would remain hydrodynamic) but could be introduced by, for example, imposing external fields or increasing drive energy to test conditions relevant to inertial confinement fusion or other high-energy-density experiments.
Here, we provide a framework to solve generic models describing the dissociation of multiple molecular Bose Einstein condensates in a nonadiabatic regime. The competition between individual chemical reactions can lead to non-trivial dependence on critical components such as path interference and symmetries, thus, affecting the final distribution of atomic population. We find an analytical solution for an illustrative example model involving four atomic modes. When the system parameters satisfy CPT symmetry, where C is charge conjugation, P is parity, and T is time-reversal symmetry, our solution predicts a population imbalance between atomic modes that is exponentially sensitive to system parameters. However, a weakly broken symmetry alters the population in each atomic mode and can reverse the population imbalance. Our solution also demonstrates a strong quantum correlation between atomic modes that leads to the spontaneous production of atoms in a multi-mode squeezed state. Moreover, in our framework, a time-dependent non-Hermitian quantum mechanics naturally manifests which can alternatively be realized experimentally in photonic systems.
The increasing scale and complexity of scientific applications are rapidly transforming the ecosystem of tools, methods, and workflows adopted by the high-performance computing (HPC) community. Big data analytics and deep learning are gaining traction as essential components in this ecosystem in a variety of scenarios, such as, steering of experimental instruments , acceleration of high-fidelity simulations through surrogate computations, and guided ensemble searches. In this context, the batch job model traditionally adopted by the supercomputing in-frastructures needs to be complemented with support to schedule opportunistic on-demand analytics jobs, leading to the problem of efficient preemption of batch jobs with minimum loss of progress. In this paper, we design and implement a simulator, CoSim, that enables on-the-fly analysis of the trade-offs arising between delaying the start of opportunistic on-demand jobs, which leads to longer analytics latency, and loss of progress due to preemption of batch jobs, which is necessary to make room for on-demand jobs. To this end, we propose an algorithm based on dynamic programming with predictable performance and scalability that enables supercomputing infrastructure schedulers to analyze the aforementioned trade-off and take decisions in near real-time. Compared with other state-of-art approaches using traces of the Theta pre-Exascale machine, our approach is capable of finding the optimal solution, while achieving high performance and scalability.
SUMMARY Plant responses to environmental change are mediated via changes in cellular metabolomes. However, <5% of signals obtained from liquid chromatography tandem mass spectrometry (LC‐MS/MS) can be identified, limiting our understanding of how metabolomes change under biotic/abiotic stress. To address this challenge, we performed untargeted LC‐MS/MS of leaves, roots, and other organs of Brachypodium distachyon (Poaceae) under 17 organ–condition combinations, including copper deficiency, heat stress, low phosphate, and arbuscular mycorrhizal symbiosis. We found that both leaf and root metabolomes were significantly affected by the growth medium. Leaf metabolomes were more diverse than root metabolomes, but the latter were more specialized and more responsive to environmental change. We found that 1 week of copper deficiency shielded the root, but not the leaf metabolome, from perturbation due to heat stress. Machine learning (ML)‐based analysis annotated approximately 81% of the fragmented peaks versus approximately 6% using spectral matches alone. We performed one of the most extensive validations of ML‐based peak annotations in plants using thousands of authentic standards, and analyzed approximately 37% of the annotated peaks based on these assessments. Analyzing responsiveness of each predicted metabolite class to environmental change revealed significant perturbations of glycerophospholipids, sphingolipids, and flavonoids. Co‐accumulation analysis further identified condition‐specific biomarkers. To make these results accessible, we developed a visualization platform on the Bio‐Analytic Resource for Plant Biology website ( https://bar.utoronto.ca/efp_brachypodium_metabolites/cgi‐bin/efpWeb.cgi ), where perturbed metabolite classes can be readily visualized. Overall, our study illustrates how emerging chemoinformatic methods can be applied to reveal novel insights into the dynamic plant metabolome and stress adaptation.
Micro- and nanomanufacturing capabilities have rapidly expanded over the past decade to include complex three-dimensional (3D) structure fabrication; however, the metrology required to accurately assess these processes via part inspection and characterization has struggled to keep pace. X-ray computed tomography (CT) is considered an ideal candidate for providing the critically needed metrology on the smallest scales, especially internal features, or inaccessible regions. X-ray CT supporting micro- and nanomanufacturing often push against the poorly understood resolution and variation limits inherent to the machines, which can distort or hide fine structures. In this study, we have developed and experimentally verify a comprehensive analytical uncertainty propagation signal variation flow graph (SVFG) model for X-ray radiography in this work to better understand resolution and image variability limits on the small scale. The SVFG approach captures, quantifies, and predicts variations occurring in the system that limit metrology capabilities, particularly in the micro/nanodomain. This work is the first step to achieving full uncertainty modeling of CT reconstructions and provides insight into improving X-ray attenuation imaging systems. The SVFG methodology framework is applied to generate a complete basis set of functions describing the major sources of variation in radiographs. Five models are identified, covering variation in energy, intensity, length, blur, and position. Radiographic system experiments are defined to measure the parameters required by the SVFGs. Best practices are identified for these measurements. The SVFG models are confirmed via direct measurement of variation to predict variation within 30% on average.
Moving toward a circular carbon economy requires enabling reuse of carbon-based macromolecules like those in biomass and polymers. Meeting this goal requires increased upcycling of waste products into higher value products. For biomass waste such as corn stover, pine needles, or bark, one approach is using pyrolysis to convert these feedstocks into bio-oil that can be used for liquid fuel or transformed into other hydrocarbon based products, like bio-polymers. Waste plastics can be upcycled or recycled into new products, or alternatively converted into bio-oil by pyrolysis. Macromolecules from biomass and polymers are challenging to convert into liquid fuel or chemical feedstocks, due to large and sometimes heterogeneous networks of polymeric bonds. We work with the Idaho National Laboratory’s Biomass Feedstock National User Facility and Department of Energy’s Feedstock Conversion Interface Consortium to develop a molecular understanding of biomass variability that occurs naturally and as a result of storage or preprocessing techniques applied to feedstocks, to understand chemical reactions occurring during pyrolysis and other biomass or polymer conversion techniques. Using two dimensional gas chromatography mass spectrometry coupled to analytical pyrolysis provides insight into the molecular changes that occur during the pyrolysis event, and these insights can extend to bench or pilot scale pyrolysis conversions. Understanding pyrolysis of biomass and polymers at a molecular level contributes to development of new tools to predict and manipulate the optimum conditions and identify the best storage, preprocessing, and conversion techniques to maximize output of specific high-value chemical species in the conversion to fuels, or to characterize feedstocks for blending before pyrolysis. Identification and understanding of the molecular level changes resulting from novel preprocessing techniques that can be used to control and manipulate the chemical output of bench or pilot scale pyrolysis is another focus of investigation, including the use of gamma irradiation, acid, or enzyme pre-treatments. We also focus on the impacts of storage conditions and techniques, which can affect moisture levels, degradation, and impacts from naturally occurring microbes in the environment.
We performed calculations of electronic excitation energies and associated oscillator strengths based on the time-dependent density-functional tight-binding (TD-DFTB) method [1]. The SMILES (Simplified molecular-input line-entry system) strings of the molecules from the AISD HOMO-LUMO database [2] were converted to a 3D atomistic structure and stored in a PDB file after preliminary geometry optimization using the Merck Molecular Force Field (MMFF94) in RDKit [3,4]. The primary information stored in the PDB file archive consists of Cartesian coordinates for each atom of the molecule in their 3D location in space, along with summary information about the structure, sequence, and experiment. We then performed molecular geometry optimization using the density-functional tight-binding (DFTB) method [5] in the electronic ground state, followed by single-point excited states calculations, as described below. We note that, since RDKit employs a random choice for the generation of molecular conformers, the molecular geometries obtained in this dataset could be different from the ones that were generated when the AISD HOMO-LUMO dataset was generated. The computed excitation energies and associated oscillator strengths can be converted to predict UV/Vis absorption spectra, where excitation energies correspond to absorption peak positions, and oscillator strengths are a good measure of the probability of absorption of visible or UV light in transitions between electronic ground and excited states. The conversion of SMILES strings to 3D Cartesian coordinates of fully DFTB-optimized molecules was successful for 10,502,904 out of 10,502,917 molecules. For these molecules, both geometry optimizations and excited states calculations were successful. The DFTB calculations did not complete for 13 molecules of the original AISD HOMO-LUMO dataset. We still provide information about the geometry of these molecules. The molecules are diverse for chemical compositions (which span 5 non-hydrogen elements: oxygen, carbon, nitrogen, fluorine, sulfur) and molecular size (the smallest molecule contains 5 non-hydrogen atoms, and the largest molecule contains 71 non-hydrogen atoms). The DFTB method [5] is an approximation to density functional theory (DFT), utilizing a minimal basis set in conjunction with a two-center approximation to the electronic Hamiltonian and overlap matrix elements. The DFTB total energy is the sum of an electronic and a repulsive energy contribution, and their calculation requires optimized electronic parameters and diatomic repulsive potential energy functions. All DFTB calculations were performed using the DFTB+ code [6] (version 21.2) and the wrapper for DFTB+ in the Atomic Simulation Environment (ASE) (version 3.22.1) [7], which performed an internal conversion of Cartesian coordinates from PDB to the .gen file format. For the geometry optimizations on the electronic ground state potential energy surface of the molecules, we have chosen the third-order DFTB (DFTB3) method [5c] and employed the matching 3ob set of electronic parameters and repulsive potentials [8]. The empirical γ-damping for hydrogen bond correction, and Grimme's D3 empirical dispersion correction with Becke-Johnson damping (D3(BJ)) [9] dispersion correction was included to improve the description of non-covalent interactions. For excited states single-point energy calculations, we employed the TD-DFTB method in conjunction with the DFTB2 method [5b] and the matching mio [5b,10] and halorg [11] parameter sets. We opted to request the simultaneous calculation of 50 excited states for singlet transition to investigate sufficient number of excited states, based on linear response theory using the Casida equation [Ref: T. A. Niehaus, S. Suhai, F. Della Sala, P Lugli, M. Elstner, G. Seifert, and Th. Frauenheim. Tight-binding approach to time-dependent density-functional response theory. Phys. Rev. B, 63:085108, 2001] and the ARPACK diagonalizer [R. B. Lehoucq, D. C. Sorensen, and C. Yang. Arpack users guide: Solution of large-scale eigenvalue problems by implicitly restarted arnoldi methods, 1997. 46, 51]. The dataset contains 1001 tar.gz files. Tar files are named as “ornl_aisd_ex_1.tar.gz†through “ornl_aisd_ex_1000.tar.gzâ€. Additionally, the 13 failed molecules are in “ornl_aisd_ex_unprocessed.tar.gzâ€. Except for the tar files listed below, each tar file contains 10,500 molecules. Tar files numbered 34, 121, 128, 352, 360, 429, 495, 509, 518, 627, 676, 668, and 862 contain 10,499 molecules each. The last tar file numbered 1000 contains 13,417 molecules. The total size of the uncompressed dataset is over 283 Gigabytes. The code for calculating the electronic excitation energies and statistical analysis of the dataset is provided at the following GitLab repository: https://github.com/ORNL/Analysis-of-Large-Scale-Molecular-Datasets-with-Python Calculating the UV spectrum of a molecule requires performing 3 main operations: 1. Converting the smiles string representation of a molecule into a geometric structure where each atom is assigned XYZ coordinates. The geometric structure is written to the file smiles.pdb. 2. Using smiles.pdb to compute the relaxed geometry of the molecule, which corresponds with the position of the atoms at the position of equilibrium at the ground state. This generates the files band.out, detailed.out, and geo_end.gen. 3. Using geo_end.gen to calculate the UV spectrum of the molecule which is written into the file EXC.DAT. Every molecule in the dataset has its own directory. The files contained in each molecule directory are as follows: 1. geo_end.gen 2. detailed.out 3. band.out 4. EXC.DAT 5. smiles.pdb REFERENCES [1] Niehaus, T. A.; Suhai, S.; Della Salla, F.; Lugli, P.; Elstner, M.; Seifert, G.; Frauenheim, Th. Tight-binding approach to time-dependent density-functional response theory. Phys. Rev. B, 2001, 63, 085108/1-9. [2] Blanchard, A.; Gounley, J.; Metha, K.; Yoo, P.; Irle, S. AISD HOMO-LUMO. DOI: 10.13139/ORNLNCCS/1869409 [3] RDKit: Cheminformatics and Machine Learning Software. 2013, [http://www.rdkit.org] [4] Tosco, P.; Stiefl, N. and Landrum, G. Bringing the MMFF force field to the RDKit: implementation and validation. J Cheminform. 2014, 6, 1–4. [5] a) Porezag, D.; Frauenheim, T.; Kohler, T.; Seifert, G.; Kaschner, Construction of tight-binding-like potentials on the basis of density-functional theory: Application to carbon, R. Phys. Rev. B 1995, 51, 12947-12957; b) Elstner, M.; Porezag, D.; Jungnickel, G.; Elsner, J.; Haugk, M.; Frauenheim, Th.; Suhai, S.; Seifert, G.; Phys. Rev. B 1998, 58, 7260-7268; c) Gaus, M.; Cui, Q.; Elstner, M. DFTB3: Extension of the Self-Consistent-Charge Density-Functional Tight-Binding Method (SCC-DFTB), J. Chem. Theory Comput. 2011, 7, 931-948; d) Cui, Q.; Elstner, M. Density functional tight binding: values of semi-empirical methods in an ab initio era, Phys. Chem. Chem. Phys. 2014, 16, 14368-14377. [6] Hourahine, B. et al. DFTB+, a software package for efficient approximate density functional theory based atomistic simulations, J. Chem. Phys. 2020, 152, 124101/1-19. [7] Larsen, A. H. et al. The atomic simulation environment—a Python library for working with atoms. J. Phys.: Cond. Matter 2017, 29, 273002. [8] Kubillus, M.; Kubar, T.; Gaus, M.; Rezac, J.; Elstner, M. Parameterization of the DFTB3 Method for Br, Ca, Cl, F, I, K, and Na in Organic and Biological Systems, J. Chem. Theory Comput. 2015, 11, 332-342. [9] Brandenburg, J. G.; Grimme, S. Accurate Modeling of Organic Molecular Crystals by Dispersion-Corrected Density Functional Tight Binding (DFTB), J. Phys. Chem. Lett. 2014, 5, 1785−1789. [10] a) Niehaus, T. A.; Elstner, M.; Frauenheim, Th.; Suhai, S. Application of an approximate density-functional method to sulfur containing compounds. J. Mol. Struct.: THEOCHEM 2001, 541, 185-94; b) Elstner, M.; Hobza, P.; Frauenheim, Th.; Suhai, S.; Kaxiras, E. Hydrogen bonding and stacking interactions of nucleic acid base pairs: A density-functional-theory based treatment. J. Chem. Phys. 2001, 114, 5149-55. [11] Kubar, T.; Bodrog, Z.; Gaus, M.; Köhler, C.; Aradi, B.; Frauenheim, Th.; Elstner, M. Parametrization of the SCC-DFTB Method for Halogens. J. Chem. Theory Comput. 2013, 9, 2939-49.
Abstract Recently, the U.S. Department of Energy (DOE), Office of Science, Biological and Environmental Research (BER), and Advanced Scientific Computing Research (ASCR) programs organized and held the Artificial Intelligence for Earth System Predictability (AI4ESP) workshop series. From this workshop, a critical conclusion that the DOE BER and ASCR community came to is the requirement to develop a new paradigm for Earth system predictability focused on enabling artificial intelligence (AI) across the field, laboratory, modeling, and analysis activities, called model experimentation (ModEx). BER’s ModEx is an iterative approach that enables process models to generate hypotheses. The developed hypotheses inform field and laboratory efforts to collect measurement and observation data, which are subsequently used to parameterize, drive, and test model (e.g., process based) predictions. A total of 17 technical sessions were held in this AI4ESP workshop series. This paper discusses the topic of the AI Architectures and Codesign session and associated outcomes. The AI Architectures and Codesign session included two invited talks, two plenary discussion panels, and three breakout rooms that covered specific topics, including 1) DOE high-performance computing (HPC) systems, 2) cloud HPC systems, and 3) edge computing and Internet of Things (IoT). We also provide forward-looking ideas and perspectives on potential research in this codesign area that can be achieved by synergies with the other 16 session topics. These ideas include topics such as 1) reimagining codesign, 2) data acquisition to distribution, 3) heterogeneous HPC solutions for integration of AI/ML and other data analytics like uncertainty quantification with Earth system modeling and simulation, and 4) AI-enabled sensor integration into Earth system measurements and observations. Such perspectives are a distinguishing aspect of this paper. Significance Statement This study aims to provide perspectives on AI architectures and codesign approaches for Earth system predictability. Such visionary perspectives are essential because AI-enabled model-data integration has shown promise in improving predictions associated with climate change, perturbations, and extreme events. Our forward-looking ideas guide what is next in codesign to enhance Earth system models, observations, and theory using state-of-the-art and futuristic computational infrastructure.