SEARCH · Engineering Papers

Results for “robust clustering”

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.

SourceResearch connectionFrom yearThrough year

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

At least 199 records · Page 11

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Non-polar ether-based electrolyte solutions for stable high-voltage non-aqueous lithium metal batteries

The electrochemical instability of ether-based electrolyte solutions hinders their practical applications in high-voltage Li metal batteries. To circumvent this issue, here, we propose a dilution strategy to lose the Li + /solvent interaction and use the dilute non-aqueous electrolyte solution in high-voltage lithium metal batteries. We demonstrate that in a non-polar dipropyl ether (DPE)-based electrolyte solution with lithium bis(fluorosulfonyl) imide salt, the decomposition order of solvated species can be adjusted to promote the Li + /salt-derived anion clusters decomposition over free ether solvent molecules. This selective mechanism favors the formation of a robust cathode electrolyte interphase (CEI) and a solvent-deficient electric double-layer structure at the positive electrode interface. When the DPE-based electrolyte is tested in combination with a Li metal negative electrode (50 μm thick) and a LiNi 0.8 Co 0.1 Mn 0.1 O 2 -based positive electrode (3.3 mAh/cm 2 ) in pouch cell configuration at 25 °C, a specific discharge capacity retention of about 74% after 150 cycles (0.33 and 1 mA/cm 2 charge and discharge, respectively) is obtained.

25 ENERGY STORAGE↗

DOE OSTI2020

Smoothers for Matrix-Free Algebraic Multigrid Preconditioning of High-Order Finite Elements

We investigate smoothers for use in matrix-free algebraic multigrid (AMG) preconditioning of high-order finite element problems. These AMG preconditioners are matrix-free in the sense that they are built from a related low-order refined finite element problem whose system matrix can be much more rapidly assembled than the high-order problem. Our proposed smoother, which we call distributive relaxation, is more robust to the anisotropy present in many low-order refined meshes which feature a clustering of nodes near the boundaries between high-order finite elements. For solving the low-order refined problem, we show that this new distributive relaxation smoother possesses significantly improved performance compared to more traditional smoothers.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

On Performance Prediction of Big Data Transfer in High-performance Networks

Big data generated by large-scale scientific and industrial applications need to be transferred between different geographical locations for remote storage, processing, and analysis. High-speed dedicated connections provisioned in High-performance Networks (HPNs) are increasingly utilized to carry out such big data transfer. HPN management highly relies on an important capability of performance (mainly throughput) prediction to reserve sufficient bandwidth and meanwhile avoid over-provisioning that may result in unnecessary resource waste. This capability is critical to improving the resource (mainly bandwidth) utilization of dedicated connections and meeting various user requests for data transfer. Conventional methods conduct performance prediction by fitting prior observed transfer history with predefined loss functions, without considering unobservable latent factors such as competing loads on end hosts. Such latent factors also have a significant impact on the application-level data transfer performance, which may result in an inaccurate prediction model. In this paper, we first investigate the impact of latent factors and propose a clustering-based method to eliminate their negative impact on performance prediction. We then develop a robust machine learning-based performance predictor by: i) incorporating the proposed latent factor elimination method into data preprocessing, and ii) adopting a customized domain guided loss function. Extensive experimental results show that our predictor achieves significantly higher prediction accuracy than several other state-of-the-art methods.

Liu, Wuji↗

DOE OSTI2025

Extending quantum-mechanical benchmark accuracy to biological ligand-pocket interactions

Predicting the binding affinity of ligands to protein pockets is key in the drug design pipeline. The flexibility of ligand-pocket motifs arises from a range of attractive and repulsive electronic interactions during binding. Accurately accounting for all interactions requires robust quantum-mechanical (QM) benchmarks, which are scarce for ligand-pocket systems. Additionally, disagreement between “gold standard” Coupled Cluster (CC) and Quantum Monte Carlo (QMC) methods casts doubt on many benchmarks for larger non-covalent systems. We introduce the “QUantum Interacting Dimer” (QUID) benchmark framework containing 170 non-covalent (non-)equilibrium systems modeling chemically and structurally diverse ligand-pocket motifs. Symmetry-adapted perturbation theory shows that QUID broadly covers non-covalent binding motifs and energetic contributions. Robust binding energies are obtained using complementary CC and QMC methods, achieving agreement of 0.5 kcal/mol. The benchmark data analysis reveals that several dispersion-inclusive density functional approximations provide accurate energy predictions, though their atomic van der Waals forces differ in magnitude and orientation. Contrarily, semiempirical methods and empirical force fields require improvements in capturing non-covalent interactions (NCIs) for out-of-equilibrium geometries. The wide span of NCIs, highly accurate interaction energies, and analysis of molecular properties take QUID beyond the “gold standard” for QM benchmarks of ligand-protein systems.

Puleva, Mirela [University of Luxembourg, Luxembou↗

DOE OSTI2023

Thermofield Theory for Finite-Temperature Electronic Structure

Wave function methods have offered a robust, systematically improvable means to study ground-state properties in quantum many-body systems. Theories like coupled cluster and their derivatives provide highly accurate approximations to the energy landscape at a reasonable computational cost. Analogues of such methods to study thermal properties, though highly desirable, have been lacking because evaluating thermal properties involve a trace over the entire Hilbert space, which is a formidable task. Besides, excited-state theories are generally not as well studied as ground-state ones. In this mini-review, we present an overview of a finite-temperature wave function formalism based on thermofield dynamics to overcome these difficulties. Thermofield dynamics allows us to map the equilibrium thermal density matrix to a pure state, i.e., a single wave function, albeit in an expanded Hilbert space. Ensemble averages become expectation values over this so-called thermal state. Around this thermal state, we have developed a procedure to generalize ground-state wave function theories to finite temperatures. As explicit examples, we highlight formulations of mean-field, configuration interaction, and coupled cluster theories for thermal properties of Fermions in the grand-canonical ensemble. To assess the quality of these approximations, we also show benchmark studies for the one-dimensional Hubbard model, while comparing against exact results. We will see that the thermal methods perform similarly to their ground-state counterparts, while merely adding a prefactor to the asymptotic computational cost. Furthermore, they also inherit all the properties, good or bad, from the ground-state methods, signifying the robustness of our formalism and the scope for future development.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Representative Period Selection for Robust Capacity Expansion Planning in Low-carbon Grids

With the increasing urgency to decarbonize power systems, while mitigating extreme events, capacity expansion models can play a vital role in reliably planning the expansion of power systems and facilitating the integration of renewable energy sources. Optimizing capacity expansion generally involves selecting surrogate representative days from forecasts of load and the generation profiles of variable renewable energy resources. To properly select those representative days, we propose a novel input-based approach in combination with the k-means clustering algorithm that utilize three unique operational inputs: load shedding, renewable curtailment, and transmission congestion. The proposed method allows for more robust and cost-effective capacity planning. The method is validated using a capacity expansion model and a production cost model based on California Independent System Operator (CAISO)'s decarbonization goals, and results in reduced costs and drastically lower load shedding.

Anderson, Osten P.↗

DOE OSTI2021

Measuring Young Stars in Space and Time. II. The Pre-main-sequence Stellar Content of N44

The Hubble Space Telescope survey Measuring Young Stars in Space and Time (MYSST) entails some of the deepest photometric observations of extragalactic star formation, capturing even the lowest-mass stars of the active star-forming complex N44 in the Large Magellanic Cloud. We employ the new MYSST stellar catalog to identify and characterize the content of young pre-main-sequence (PMS) stars across N44 and analyze the PMS clustering structure. To distinguish PMS stars from more evolved line of sight contaminants, a non-trivial task due to several effects that alter photometry, we utilize a machine-learning classification approach. This consists of training a support vector machine (SVM) and a random forest (RF) on a carefully selected subset of the MYSST data and categorize all observed stars as PMS or non-PMS. Combining SVM and RF predictions to retrieve the most robust set of PMS sources, we find ∼26,700 candidates with a PMS probability above 95% across N44. Employing a clustering approach based on a nearest neighbor surface density estimate, we identify 16 prominent PMS structures at 1σ significance above the mean density with sub-clusters persisting up to and beyond 3σ significance. The most active star-forming center, located at the western edge of N44's bubble, is a subcluster with an effective radius of ∼5.6 pc entailing more than 1100 PMS candidates. Furthermore, we confirm that almost all identified clusters coincide with known H ii regions and are close to or harbor massive young O stars or YSOs previously discovered by MUSE and Spitzer observations.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2021

Exploratory analysis and performance prediction of big data transfer in High-performance Networks

Big data transfer in large-scale scientific and business applications is increasingly carried out over connections with guaranteed bandwidth provisioned in High-performance Networks (HPNs) via advance bandwidth reservation. Provisioning agents need to carefully schedule data transfer requests, compute network paths, and allocate appropriate bandwidths. Such reserved bandwidths, if not fully utilized, could be simply wasted due to the exclusive access during the approved time window, and cause extra overhead and complexity for resource management. This calls for accurate performance prediction to reserve bandwidths that match actual needs and avoid over-provisioning. We employ machine learning algorithms to predict big data transfer performance based on extensive performance measurements collected in the past several years from data transfer tests using different protocols and toolkits between various end sites on several real-life physical or emulated testbeds. We first analyze the performance patterns in response to a comprehensive list of parameters in end-host systems, network connections, and data transfer applications, which motivate the use of machine learning and also help us identify the effects of latent factors. We then propose threshold- and clustering-based methods to eliminate negative effects of latent factors in data preprocessing and build a robust performance predictor based on customized domain-oriented loss functions. The performance of the proposed methods is verified by extensive experiments using SVR and RFR as well as theoretical analysis of the general performance bound.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Comparing Compressed and Full-Modeling analyses with FOLPS: implications for DESI 2024 and beyond

The Dark Energy Spectroscopic Instrument (DESI) will provide unprecedented information about the large-scale structure of our Universe. In this work, we study the robustness of the theoretical modelling of the power spectrum of F OLPS , a novel effective field theory-based package for evaluating the redshift space power spectrum in the presence of massive neutrinos. We perform this validation by fitting the AbacusSummit high-accuracy N -body simulations for Luminous Red Galaxies, Emission Line Galaxies and Quasar tracers, calibrated to describe DESI observations. We quantify the potential systematic error budget of F OLPS finding that the modelling errors are fully sub-dominant for the DESI statistical precision within the studied range of scales. Additionally, we study two complementary approaches to fit and analyse the power spectrum data, one based on direct Full-Modelling fits and the other on the ShapeFit compression variables, both resulting in very good agreement in precision and accuracy. In each of these approaches, we study a set of potential systematic errors induced by several assumptions, such as the choice of template cosmology, the effect of prior choice in the nuisance parameters of the model, or the range of scales used in the analysis. Furthermore, we show how opening up the parameter space beyond the vanilla ΛCDM model affects the DESI observables. These studies include the addition of massive neutrinos, spatial curvature, and dark energy equation of state. We also examine how relaxing the usual Cosmic Microwave Background and Big Bang Nucleosynthesis priors on the primordial spectral index and the baryonic matter abundance, respectively, impacts the inference on the rest of the parameters of interest. This paper pathways towards performing a robust and reliable analysis of the shape of the power spectrum of DESI galaxy and quasar clustering using F OLPS .

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

Exploiting non-linear scales in galaxy–galaxy lensing and galaxy clustering: A forecast for the dark energy survey

ABSTRACT The combination of galaxy–galaxy lensing (GGL) and galaxy clustering is a powerful probe of low-redshift matter clustering, especially if it is extended to the non-linear regime. To this end, we use an N-body and halo occupation distribution (HOD) emulator method to model the redMaGiC sample of colour-selected passive galaxies in the Dark Energy Survey (DES), adding parameters that describe central galaxy incompleteness, galaxy assembly bias, and a scale-independent multiplicative lensing bias Alens. We use this emulator to forecast cosmological constraints attainable from the GGL surface density profile ΔΣ(rp) and the projected galaxy correlation function wp, gg(rp) in the final (Year 6) DES data set over scales $r_p=0.3\!-\!30.0\, h^{-1} \, \mathrm{Mpc}$. For a $3{{\ \rm per\ cent}}$ prior on Alens we forecast precisions of $1.9{{\ \rm per\ cent}}$, $2.0{{\ \rm per\ cent}}$, and $1.9{{\ \rm per\ cent}}$ on Ωm, σ8, and $S_8 \equiv \sigma _8\Omega _m^{0.5}$, marginalized over all halo occupation distribution (HOD) parameters as well as Alens. Adding scales $r_p=0.3\!-\!3.0\, h^{-1} \, \mathrm{Mpc}$ improves the S8 precision by a factor of ∼1.6 relative to a large scale ($3.0\!-\!30.0\, h^{-1} \, \mathrm{Mpc}$) analysis, equivalent to increasing the survey area by a factor of ∼2.6. Sharpening the Alens prior to $1{{\ \rm per\ cent}}$ further improves the S8 precision to $1.1{{\ \rm per\ cent}}$, and it amplifies the gain from including non-linear scales. Our emulator achieves per cent-level accuracy similar to the projected DES statistical uncertainties, demonstrating the feasibility of a fully non-linear analysis. Obtaining precise parameter constraints from multiple galaxy types and from measurements that span linear and non-linear clustering offers many opportunities for internal cross-checks, which can diagnose systematics and demonstrate the robustness of cosmological results.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2023

The use of double-mode RR Lyrae stars as robust distance and metallicity indicators

Abstract RR Lyrae stars are one of the primary distance indicators for old stellar populations such as globular clusters, dwarf galaxies and galaxies. Typically, fundamental-mode RR Lyr stars are used for distance measurements, and their accuracy is strongly limited by the dependence of absolute magnitudes on metallicity, in both the optical and infrared bands. Here, we report the discovery of a period–(period ratio)–metallicity relation for double-mode RR Lyr stars, which can predict metallicity as accurately as the low-resolution spectra. With theoretical and observational evidence, we propose that the period–luminosity relation of double-mode RR Lyr stars is not affected by the metallicity. Combining the Large Magellanic Cloud distance and Gaia parallaxes, we calibrate the zero point of the period–luminosity relation to an error of 0.022 mag, which means that in the best case double-mode RR Lyr stars can anchor galaxy distances to an accuracy of 1.0%. For four globular clusters and two dwarf galaxies, we obtain distances using double-mode RR Lyr stars with a distance accuracy of 2–3% and 1–2%, respectively. With future telescopes such as the China Space Station Telescope and the Vera C. Rubin Observatory, double-mode RR Lyr stars will be established as an independent distance ladder in the near-field universe.

Astronomy & Astrophysics↗

DOE OSTI2023

An accurate and efficient fragmentation approach via the generalized many-body expansion for density matrices

With relevant chemical space growing larger and larger by the day, the ability to extend computational tractability over that larger space is of paramount importance in virtually all fields of science. The solution we aim to provide here for this issue is in the form of the generalized many-body expansion for building density matrices (GMBE-DM) based on the set-theoretical derivation with overlapping fragments, through which the energy can be obtained by a single Fock build. In combination with the purification scheme and the truncation at the one-body level, the DM-based GMBE(1)-DM-P approach shows both highly accurate absolute and relative energies for medium-to-large size water clusters with about an order of magnitude better than the corresponding energy-based GMBE(1) scheme. Simultaneously, GMBE(1)-DM-P is about an order of magnitude faster than the previously proposed MBE-DM scheme [F. Ballesteros and K. U. Lao, J. Chem. Theory Comput. 18, 179 (2022)] and is even faster than a supersystem calculation without significant parallelization to rescue the fragmentation method. For even more challenging systems including ion–water and ion–pair clusters, GMBE(1)-DM-P also performs about 3 and 30 times better than the energy-based GMBE(1) approach, respectively. In addition, this work provides the first overlapping fragmentation algorithm with a robust and effective binning scheme implemented internally in a popular quantum chemistry software package. Thus, GMBE(1)-DM-P opens a new door to accurately and efficiently describe noncovalent clusters using quantum mechanics.

Chemistry↗

DOE OSTI2022

Dark Energy Survey Year 3 results: Cosmology from combined galaxy clustering and lensing validation on cosmological simulations

In this work, we present a validation of the Dark Energy Survey Year 3 (DES Y3) 3×2-point analysis choices by testing them on Buzzard2.0, a new suite of cosmological simulations that is tailored for the testing and validation of combined galaxy clustering and weak-lensing analyses. We show that the buzzard2.0 simulations accurately reproduce many important aspects of the DES Y3 data, including photometric redshift and magnitude distributions, and the relevant set of two-point clustering and weak-lensing statistics. We then show that our model for the 3×2-point data vector is accurate enough to recover the true cosmology in simulated surveys assuming the true redshift distributions for our source and lens samples, demonstrating robustness to uncertainties in the modeling of the nonlinear matter power spectrum, nonlinear galaxy bias, and higher-order lensing corrections. Additionally, we demonstrate for the first time that our photometric redshift calibration methodology, including information from photometry, spectroscopy, clustering cross-correlations, and galaxy–galaxy lensing ratios, is accurate enough to recover the true cosmology in simulated surveys in the presence of realistic photometric redshift uncertainties.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2004

Functional Groups Based on Leaf Physiology: Are they Spatially and Temporally Robust?

The functional grouping hypothesis, which suggests that complexity in ecosystem function can be simplified by grouping species with similar responses, was tested in the Florida scrub habitat. Functional groups were identified based on how species in fire maintained Florida scrub regulate exchange of carbon and water with the atmosphere as indicated by both instantaneous gas exchange measurements and integrated measures of function (%N, delta C-13, delta N-15, C-N ratio). Using cluster analysis, five distinct physiologically-based functional groups were identified in the fire maintained scrub. These functional groups were tested to determine if they were robust spatially, temporally, and with management regime. Analysis of Similarities (ANOSIM), a non-parametric multivariate analysis, indicated that these five physiologically-based groupings were not altered by plot differences (R = -0.115, p = 0.893) or by the three different management regimes; prescribed burn, mechanically treated and burn, and fire-suppressed (R = 0.018, p = 0.349). The physiological groupings also remained robust between the two climatically different years 1999 and 2000 (R = -0.027, p = 0.725). Easy-to-measure morphological characteristics indicating functional groups would be more practical for scaling and modeling ecosystem processes than detailed gas-exchange measurements, therefore we tested a variety of morphological characteristics as functional indicators. A combination of non-parametric multivariate techniques (Hierarchical cluster analysis, non-metric Multi-Dimensional Scaling, and ANOSIM) were used to compare the ability of life form, leaf thickness, and specific leaf area classifications to identify the physiologically-based functional groups. Life form classifications (ANOSIM; R = 0.629, p 0.001) were able to depict the physiological groupings more adequately than either specific leaf area (ANOSIM; R = 0.426, p = 0.001) or leaf thickness (ANOSIM; R 0.344, p 0.001). The ability of life forms to depict the physiological groupings was improved by separating the parasitic Ximenia americana from the shrub category (ANOSIM; R = 0.794, p = 0.001). Therefore, a life form classification including parasites was determined to be a good indicator of the physiological processes of scrub species, and would be a useful method of grouping for scaling physiological processes to the ecosystem level.

Foster, Tammy E.↗

NASA NTRS1995

The coupling of fluids, dynamics, and controls on advanced architecture computers

This grant provided for the demonstration of coupled controls, body dynamics, and fluids computations in a workstation cluster environment; and an investigation of the impact of peer-peer communication on flow solver performance and robustness. The findings of these investigations were documented in the conference articles.The attached publication, 'Towards Distributed Fluids/Controls Simulations', documents the solution and scaling of the coupled Navier-Stokes, Euler rigid-body dynamics, and state feedback control equations for a two-dimensional canard-wing. The poor scaling shown was due to serialized grid connectivity computation and Ethernet bandwidth limits. The scaling of a peer-to-peer communication flow code on an IBM SP-2 was also shown. The scaling of the code on the switched fabric-linked nodes was good, with a 2.4 percent loss due to communication of intergrid boundary point information. The code performance on 30 worker nodes was 1.7 (mu)s/point/iteration, or a factor of three over a Cray C-90 head. The attached paper, 'Nonlinear Fluid Computations in a Distributed Environment', documents the effect of several computational rate enhancing methods on convergence. For the cases shown, the highest throughput was achieved using boundary updates at each step, with the manager process performing communication tasks only. Constrained domain decomposition of the implicit fluid equations did not degrade the convergence rate or final solution. The scaling of a coupled body/fluid dynamics problem on an Ethernet-linked cluster was also shown.

Atwood, Christopher↗

DOE OSTI2026

Reconstructing Quasar Spectra and Measuring the Lyα Forest with SpenderQ

Quasar spectra carry the imprint of foreground intergalactic medium (IGM) through absorption features. In particular, absorption caused by neutral hydrogen gas, the "Lyα forest," is a key spectroscopic tracer for cosmological analyses used to measure cosmic expansion and test physics beyond the standard model. Despite their importance, current methods for measuring Lyα absorption cannot directly derive the intrinsic quasar continuum and make strong assumptions on its shape, thus distorting the measured Lyα clustering. We present SpenderQ , a ML-based approach for directly reconstructing the intrinsic quasar spectra and measuring the Lyα forest from observations. SpenderQ uses the Spender spectrum autoencoder to learn a compact and redshift-invariant latent encoding of quasar spectra, combined with an iterative procedure to identify and mask absorption regions. To demonstrate its performance, we apply SpenderQ to 400,000 synthetic quasar spectra created to validate the Dark Energy Spectroscopic Instrument Year 1 Lyα cosmological analyses. SpenderQ accurately reconstructs the true intrinsic quasar spectra, including the broad Lyβ, Lyα, SiIV, CIV, and CIII emission lines. Redward of Lyα, SpenderQ provides percent-level reconstructions of the true quasar spectra. Blueward of Lyα, SpenderQ reconstructs the true spectra to < 5%. SpenderQ reproduces the shapes of individual quasar spectra more robustly than the current state-of-the-art. We, thus, expect it will significantly reduce biases in Lyα clustering measurements and enable studies of quasars and their physical properties. SpenderQ also provides informative latent variable encodings that can be used to, e.g., classify quasars with Broad Absorption Lines. Overall, SpenderQ provides a new data-driven approach for unbiased Lyα forest measurements in cosmological, quasar, and IGM studies.

Hahn, ChangHoon [Arizona U., Astron. Dept. - Stewa↗

DOE OSTI2023

GAHLS: an optimized graph analytics based high level synthesis framework

The urgent need for low latency, high-compute and low power on-board intelligence in autonomous systems, cyber-physical systems, robotics, edge computing, evolvable computing, and complex data science calls for determining the optimal amount and type of specialized hardware together with reconfigurability capabilities. With these goals in mind, we propose a novel comprehensive graph analytics based high level synthesis (GAHLS) framework that efficiently analyzes complex high level programs through a combined compiler-based approach and graph theoretic optimization and synthesizes them into message passing domain-specific accelerators. This GAHLS framework first constructs a compiler-assisted dependency graph (CaDG) from low level virtual machine (LLVM) intermediate representation (IR) of high level programs and converts it into a hardware friendly description representation. Next, the GAHLS framework performs a memory design space exploration while account for the identified computational properties from the CaDG and optimizing the system performance for higher bandwidth. The GAHLS framework also performs a robust optimization to identify the CaDG subgraphs with similar computational structures and aggregate them into intelligent processing clusters in order to optimize the usage of underlying hardware resources. Finally, the GAHLS framework synthesizes this compressed specialized CaDG into processing elements while optimizing the system performance and area metrics. Evaluations of the GAHLS framework on several real-life applications (e.g., deep learning, brain machine interfaces) demonstrate that it provides 14.27× performance improvements compared to state-of-the-art approaches such as LegUp 6.2.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Topological grain boundary segregation transitions

Engineering the structure of grain boundaries (GBs) by solute segregation is a promising strategy to tailor the properties of polycrystalline materials. Solute segregation triggering phase transitions at GBs has been suggested theoretically to offer different pathways to design interfaces, but an understanding of their intrinsic atomistic nature is missing. Here, we combined atomic resolution electron microscopy and atomistic simulations to discover that iron segregation to GBs in titanium stabilizes icosahedral units (“cages”) that form robust building blocks of distinct GB phases. Owing to their five-fold symmetry, the iron cages cluster and assemble into hierarchical GB phases characterized by a different number and arrangement of the constituent icosahedral units. Our advanced GB structure prediction algorithms and atomistic simulations validate the stability of these observed phases and the high excess of iron at the GB that is accommodated by the phase transitions.

36 MATERIALS SCIENCE↗