Search NASA⌕ Search

Engineering topics

Park, Haesun

Publications and source records attributed to Park, Haesun.

At least 19 records

Entropy-Stabilized Multication Fluorides as a Conversion-Type Cathode for Li-Ion Batteries–Impact of Element Selection

Metal fluorides (e.g., FeF 2 and FeF 3 ) have received attention as conversion-type cathode materials for Li-ion batteries due to their higher theoretical capacity compared to that of common intercalation materials. However, their practical use has been hindered by low round-trip efficiency, voltage hysteresis, and capacity fading. Cation substitution has been proposed to address these challenges, and recent advancements in battery performance involve the introduction of entropy stabilization in an attempt to facilitate reversible conversion reactions by increasing configurational entropy. Building on this concept, high entropy fluorides with five cations were synthesized by using a simple mechanochemical route. In order to examine the impact of element selection, Co 0.2 Cu 0.2 Ni 0.2 Zn 0.2 Fe 0.2 F 2 (HEF-Fe) was compared with Co 0.2 Cu 0.2 Ni 0.2 Zn 0.2 Mg 0.2 F 2 (HEF-Mg), replacing electrochemically inactive Mg with Fe as an active participant in the conversion reaction. Combining electrochemical measurements with first-principles calculations, high-resolution electron microscopy, and synchrotron X-ray analysis, HEFs’ battery performances and conversion reaction mechanisms were investigated in detail. The results highlighted that replacement of Mg with Fe was beneficial, with enhanced capacity, rate capability, and surface stability. In addition, it was found that HEF-Fe showed similar cycle stability without an electrochemically inactive element. In conclusion, these findings provide valuable insights for the design of high entropy multielement fluorides for improved Li-ion battery performance.

25 ENERGY STORAGE↗

Strain-Driven Surface Reactivity in Magnesium-Ion Battery Cathodes

Successful deployment of a Mg-ion battery requires cathodes that can achieve reversible Mg intercalation and high energy density. Recent theoretical and experimental studies indicated that the overall transport is likely limited by sluggish Mg transport at the cathode-electrolyte interface and not Mg diffusion through bulk. Here, in this work, we investigated the surface electrochemical activity of Mg ions by using a spinel-structured manganese oxide thin-film model system and in situ X-ray scattering. In combination with post-mortem microscopy analysis, we found that magnesium insertion was more favorable than subsequent extraction near the surface of the Mg x Mn 2 O 4 film, resulting in overmagnesiation, and eventually amorphization of the surface. This structural irreversibility and high overpotential required for Mg extraction could explain significant voltage hysteresis and Mg surface enrichment previously observed in bulk cathodes. Density functional theory calculations suggested that the tendency for the Mg surface enrichment could be associated with Mg diffusion kinetics, which varies with the strain state evolved due to constrained film volume change during Mg insertion and extraction. Particularly, out-of-plane Mg migration was predicted to be favorable in the tensile strain rather than in the compressive case.

25 ENERGY STORAGE↗

Distributed-Memory Parallel JointNMF

Joint Nonnegative Matrix Factorization (JointNMF) is a hybrid method for mining information from datasets that contain both feature and connection information. We propose distributed-memory parallelizations of three algorithms for solving the JointNMF problem based on Alternating Nonnegative Least Squares, Projected Gradient Descent, and Projected Gauss-Newton. We extend well-known communication-avoiding algorithms using a single processor grid case to our coupled case on two processor grids. We demonstrate the scalability of the algorithms on up to 960 cores (40 nodes) with 60% parallel efficiency. The more sophisticated Alternating Nonnegative Least Squares (ANLS) and Gauss-Newton variants outperform the first-order gradient descent method in reducing the objective on large-scale problems. We perform a topic modelling task on a large corpus of academic papers that consists of over 37 million paper abstracts and nearly a billion citation relationships, demonstrating the utility and scalability of the methods.

Eswar, Srinivas↗

Investigation of Rechargeable Calcium Metal-Selenium Batteries Enabled by Borate-Based Electrolytes

Calcium-ion batteries (CIBs) are a promising next-generation energy storage system given the low redox potential of calcium metal and high abundance of calcium compounds. For continued CIB development, the discovery of high energy density calcium ion cathodes is needed to achieve practical energy density values. Here, we report on the use of elemental Se as a promising candidate for a high-capacity cathode material for CIBs that operates via a conversion mechanism in a Ca metal battery at room temperature. The Se electrodes demonstrate a reversible specific capacity of 180 mA h g –1 with a discharge plateau near 2.0 V (vs Ca 2+ /Ca) at 100 mA g –1 using an electrolyte based on the salt calcium tetrakis(hexafluoroisopropyloxy)borate (Ca(B(hfip) 4 ) 2 ) in 1,2-dimethoxyethane (DME) and Ca metal. The reversible electrochemical reaction between calcium and selenium is investigated using operando synchrotron-based techniques and the possible reaction mechanism discussed.

25 ENERGY STORAGE↗

Skew-Symmetric adjacency matrices for clustering directed graphs

Graph clustering methods often critically rely on the symmetry of graph matrices. Developing analogous methods for digraphs often proves more challenging, because digraph matrices are typically asymmetric and not orthogonally diagonalizable. However, researchers have recently proposed several complex-valued Hermitian digraph matrices. In particular, one such representation has been utilized as an input to an algorithm for finding imbalanced cuts. In this work, we establish an algebraic relationship between this matrix and an associated real-valued matrix. We show using this real-valued matrix for imbalanced cut-finding algorithms is not only sufficient but advantageous. Our algorithm uses less memory and asymptotically less computation while provably preserving solution quality. We also show our method can be easily implemented using standing computational building blocks, possesses better numerical properties, and loans itself to a natural interpretation via an objective function relaxation argument. We empirically demonstrate these advantages on real world data sets and show how our algorithm can uncover meaningful cluster structure.

Hayashi, Koby↗

Investigation of Ca Insertion into α-MoO 3 Nanoparticles for High Capacity Ca-Ion Cathodes

Calcium-ion batteries (CIBs) are a promising alternative to lithium-ion batteries (LIBs) due to the low redox potential of calcium metal and high abundance of calcium compounds. Due to its layered structure, α-MoO 3 is regarded as a promising cathode host lattice. While studies have reported that α-MoO 3 can reversibly intercalate Ca ions, limited electrochemical activity has been noted, and its reaction mechanism remains unclear. Here, we re-examine Ca insertion into α-MoO 3 nanoparticles with a goal to improve reaction kinetics and clarify the storage mechanism. The α-MoO 3 electrodes demonstrated a specific capacity of 165 mA h g –1 centered near 2.7 V vs Ca 2+ /Ca, stable long-term cycling, and good rate performance at room temperature. Furthermore, this work demonstrates that, under the correct conditions, layered oxides can be a promising host material for CIBs and renews prospects for CIBs.

36 MATERIALS SCIENCE↗

Layered Transition Metal Oxides as Ca Intercalation Cathodes: A Systematic First-Principles Evaluation

Finding high-voltage Ca cathode materials is a critical step to unleashing the full potential of high-energy-density Ca-ion batteries. Therefore, first-principles calculations are used to demonstrate that P-type layered calcium transition metal (TM) oxide materials (CaTM 2 O 4 ) with a range of TM substitutions (TM = Ti, V, Cr, Mn, Fe, Co, and Ni) have excellent battery-related properties including thermodynamic stability, average voltage, energy density, synthesizability, ionic mobility, and electronic structure. However, the thermodynamic stability of the charged phase and TM redox activity are shown to be sensitive to TM selection, with CaCo 2 O 4 having the best balance of all considered properties. The utility of combining multiple TMs to expand the chemical search space for TM substitutions is demonstrated by mixing Co and Ni in layered CaTM 2 O 4 .

25 ENERGY STORAGE↗

Atomic-Scale Structure of Chemically Distinct Surface Oxygens in Redox Reactions

During redox reactions, oxide-supported catalytic systems undergo structural and chemical changes. Improving subsequent catalytic properties requires an understanding of the atomic-scale structure with chemical state specificity under reaction conditions. For the case of 1/2 monolayer vanadia on α-TiO 2 (110), we use X-ray standing wave (XSW) excited X-ray photoelectron spectroscopy to follow the redox induced atomic positional and chemical state changes of this interface. Furthermore, while the resulting XSW 3D composite atomic maps include the Ti and O substrate atoms and V surface atoms, our focus in this report is on the previously unseen surface oxygen species with comparison to density functional theory predictions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

PLANC: Parallel Low-rank Approximation with Nonnegativity Constraints

In this work, we consider the problem of low-rank approximation of massive dense nonnegative tensor data, for example, to discover latent patterns in video and imaging applications. As the size of data sets grows, single workstations are hitting bottlenecks in both computation time and available memory. We propose a distributed-memory parallel computing solution to handle massive data sets, loading the input data across the memories of multiple nodes, and performing efficient and scalable parallel algorithms to compute the low-rank approximation. We present a software package called Parallel Low-rank Approximation with Nonnegativity Constraints, which implements our solution and allows for extension in terms of data (dense or sparse, matrices or tensors of any order), algorithm (e.g., from multiplicative updating techniques to alternating direction method of multipliers), and architecture (we exploit GPUs to accelerate the computation in this work). We describe our parallel distributions and algorithms, which are careful to avoid unnecessary communication and computation, show how to extend the software to include new algorithms and/or constraints, and report efficiency and scalability results for both synthetic and real-world data sets.

97 MATHEMATICS AND COMPUTING↗

ORCA: Outlier detection and Robust Clustering for Attributed graphs

Here, a framework is proposed to simultaneously cluster objects and detect anomalies in attributed graph data. Our objective function along with the carefully constructed constraints promotes interpretability of both the clustering and anomaly detection components, as well as scalability of our method. In addition, we developed an algorithm called Outlier detection and Robust Clustering for Attributed graphs (ORCA) within this framework. ORCA is fast and convergent under mild conditions, produces high quality clustering results, and discovers anomalies that can be mapped back naturally to the features of the input data. The efficacy and efficiency of ORCA is demonstrated on real world datasets against multiple state-of-the-art techniques.

97 MATHEMATICS AND COMPUTING↗

Direct Observation of Electron Beam-Induced Phase Transition in MgCrMnO 4

Irreversible structural transformation in intercalation-type cathode materials, which has been frequently observed, has been perceived as a principal cause of capacity fading and voltage decay in (uni) multivalent batteries. Herein, we explored the electron beam-induced spinel to defective rocksalt phase transitions in MgCrMnO 4 , a potential multivalent cation intercalation cathode, using atomic-resolution imaging and spectroscopy in an aberration-corrected scanning transmission electron microscope. This dynamic electron beam irradiation study of specific structural transformations provides an atomistic understanding of the structural evolution observed in transition-metal oxide spinels during electrochemical cycling using multivalent cations, such as Mg 2+ . By combining an imaging study with first-principles modeling, we demonstrate that the mechanism of the spinel to defective rocksalt transformation in MgCrMnO 4 nanostructures is enabled by the presence of oxygen vacancies and is, therefore, very similar to that observed in transition-metal oxide spinels upon Li intercalation.

36 MATERIALS SCIENCE↗

Distributed-Memory Parallel Symmetric Nonnegative Matrix Factorization

We develop the first distributed-memory parallel implementation of Symmetric Nonnegative Matrix Factorization (SymNMF), a key data analytics kernel for clustering and dimensionality reduction. Our implementation includes two different algorithms for SymNMF, which give comparable results in terms of time and accuracy. The first algorithm is a parallelization of an existing sequential approach that uses solvers for non symmetric NMF. The second algorithm is a novel approach based on the Gauss-Newton method. It exploits second-order information without incurring large computational and memory costs. We evaluate the scalability of our algorithms on the Summit system at Oak Ridge National Laboratory, scaling up to 128 nodes (4,096 cores) with 70% efficiency. Additionally, we demonstrate our software on an image segmentation task.

Eswar, Srinivas↗

Hypergraph Random Walks, Laplacians, and Clustering

We propose a flexible framework for clustering hypergraph-structured data based on recently proposed random walks utilizing edge-dependent vertex weights. When incorporating edge-dependent vertex weights (EDVW), a weight is associated with each vertex-hyperedge pair, yielding a weighted incidence matrix of the hypergraph. Such weightings have been utilized in term-document representations of text data sets. We explain how random walks with EDVW serve to construct different hypergraph Laplacian matrices, and then develop a suite of clustering methods that use these incidence matrices and Laplacians for hypergraph clustering. Using 20Newsgroup, U.S. patent, Reuters' Corpus Volume 1, and genetics data sets, we compare the performance of these clustering algorithms experimentally against a variety of existing hypergraph clustering methods. We show that the proposed methods produce higher-quality clusters.

hypergraphs, clustering, laplacian, random walk, M↗

Thermodynamic and kinetic properties of layered-CaCo2O4 for the Ca-ion batteries: a systematic first-principles study

One of the more promising directions in multivalent energy storage is systems based on Ca ion intercalation due to the potential for high voltage and capacity. A major challenge for enabling such a battery is to find cathode materials capable of fast ionic diffusion and reversible insertion of Ca ions. Here, on the basis of first-principles calculations, we have demonstrated that layered CaCo 2 O 4 exhibits favorable thermodynamic and kinetic properties that should enable topotactic Ca ion intercalation reactions. The P3-type layered Ca x Co 2 O 4 (0 < x < 1) with either of space groups of P 1 or P 2 1 / m are stable at multiple Ca concentrations and show a smooth voltage plateau higher than 3 V up to X = 0.5. The energy barriers of the single Ca ion migration are as low as 0.36 eV and 0.27 eV at the dilute and high vacancy concentration limits, respectively. Therefore, although varying the vacancy environments of the diffusing atom influences the migration barriers, they do not exceed 0.6 eV. Stochastic analysis of Ca hopping events performed by ab initio molecular dynamics (AIMD) simulation has shown that the migration barriers are lower than 0.32 eV. Therefore, the Ca diffusivity at room temperature extrapolated from the AIMD results is comparable to Li diffusivity (>10 -10 cm 2 s -1 ) in conventional Li cathode materials, suggesting the feasibility of layered Ca x Co 2 O 4 as multivalent cathode materials. Finally, the structural factors that enable fast diffusion are discussed.

25 ENERGY STORAGE↗

High-Voltage Phosphate Cathodes for Rechargeable Ca-Ion Batteries

Calcium-ion batteries (CIBs) are under investigation as next-generation energy storage devices due to their theoretically high operating potentials and lower costs tied to the high natural abundance of calcium. However, the development of CIBs has been limited by the lack of available positive electrode materials. In this study, for the first time, we report two functional polyanionic phosphate materials as high-voltage cathodes for CIBs at room temperature. NaV 2 (PO 4 ) 3 electrodes were found to reversibly intercalate 0.6 mol of Ca 2+ (81 mA h g –1 ) near 3.2 V (vs Ca 2+ /Ca) with stable cycling performance at a current density of 3.5 mA g –1 . The olivine framework material FePO 4 reversibly intercalates 0.2 mol of Ca 2+ (72 mA h g –1 ) near 2.9 V (vs Ca 2+ /Ca) at a current density of 7.5 mA g –1 in the first cycle. Structural, electronic, and compositional changes are consistent with reversible Ca 2+ intercalation into these two materials.

25 ENERGY STORAGE↗

High Capacity for Mg 2+ Deintercalation in Spinel Vanadium Oxide Nanocrystals

Nonaqueous Mg batteries can theoretically reach high energy density with cost-effective materials, yet no such device to date has performance competitive with Li-ion technologies. A major barrier is the need for oxide cathodes that combine high capacity and voltage. Very few oxides have shown intrinsic ability for Mg 2+ intercalation in electrolytes with acceptably low content of H2O. Herein, we demonstrate that nanocrystals of MgV 2 O 4 can reach high capacity for Mg 2+ deintercalation with a mechanism that preserves their spinel framework, validated through measurements with different chemical and structural sensitivity. The structural stability contrasts with other phases where reaching high capacity required distortions that introduce undesirable mechanical strain. The favorable properties of the oxide allowed cycling in a full cell with Mg metal. This work reveals new insights into the viability of multivalent intercalation in oxides, meeting a milestone toward the feasibility of high-voltage batteries with either Mg metal or Mg-ion anodes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

High Voltage Mg-Ion Battery Cathode via a Solid Solution Cr–Mn Spinel Oxide

We discuss how lattice Mg 2+ in a tailored solid solution spinel, MgCrMnO 4 , is electrochemically utilized at high Mn-redox potentials in a nonaqueous electrolyte. Complementary evidence from experimental and theoretical analyses supports bulk Mg 2+ (de)intercalation throughout the designed oxide frame where strong electrostatic interaction between Mg 2+ and O 2- exists. Mg/Mn antisite inversion in the spinel is lowered to similar to 10% via postannealing at 350 degrees C to further improve Mg 2+ mobility. Spinel lattice is preserved upon removal of Mg 2+ without any phase transformations, denoting structural stability at the charged state at a high potential similar to 3.0 V (vs Mg/Mel. Clear remagnesiation upon first discharge, harvesting up to similar to 180 Wh/kg at 60 degrees C is shown. In the remagnesiated state, insertion of Mg 2+ into interstitial sites in the spinel is detected, possibly resulting in partial reversibility which needs to be addressed for structural stability. The observations constitute a first clear path to the development of a practical high voltage Mg-ion cathode using a spinel oxide.

25 ENERGY STORAGE↗