Shashwat Anand
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Shashwat Anand has a total of 104 co-authors. Here are the top co-authors by number of shared publications:
- Gerald Jeffrey Snyder
Northwestern University
19 shared publications - Gerbrand Ceder
Department of Materials Science and Engineering University of California Berkeley California USA
6 shared publications - James Patrick Male
Materials Science and Engineering, Northwestern University, Evanston, Illinois 60208, United States
5 shared publications - Chris Wolverton
International Institute for Nanotechnology Northwestern University Evanston Illinois USA
3 shared publications Kaiyang Xia
State Key Laboratory of Silicon Materials, School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027, China
3 shared publications- Kazuki Imasato
Global Zero Emission Research Center
3 shared publications
Latest Publications
- 17 July 2025
- Published by Elsevier in Materials Today
- Vol. 88, 210-218
Origin of enhanced performance when Mn-rich rocksalt cathodes transform to δ-DRX
- 17 July 2025
- Published by Elsevier in Materials Today
- Vol. 88, 210-218
Origin of enhanced performance when Mn-rich rocksalt cathodes transform to δ-DRX
- Top Cited PapersOpen Access
- 10 January 2025
- Published by Springer Nature in npj Computational Materials
- Vol. 11 (1) , 1-9
Systematic softening in universal machine learning interatomic potentials
Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have led to universal MLIPs (uMLIPs) that are pre-trained on diverse datasets, providing opportunities for universal force fields and foundational machine learning models. However, their performance in extrapolating to out-of-distribution complex atomic environments remains unclear. In this study, we highlight a consistent potential energy surface (PES) softening effect in three uMLIPs: M3GNet, CHGNet, and MACE-MP-0, which is characterized by energy and force underprediction in atomic-modeling benchmarks including surfaces, defects, solid-solution energetics, ion migration barriers, phonon vibration modes, and general high-energy states. The PES softening behavior originates primarily from the systematically underpredicted PES curvature, which derives from the biased sampling of near-equilibrium atomic arrangements in uMLIP pre-training datasets. Our findings suggest that a considerable fraction of uMLIP errors are highly systematic, and can therefore be efficiently corrected. We argue for the importance of a comprehensive materials dataset with improved PES sampling for next-generation foundational MLIPs.
- 10 January 2025
- Published by Springer Nature in npj Computational Materials
- Vol. 11 (1) , 1-9
Top Cited PapersOpen AccessSystematic softening in universal machine learning interatomic potentials
Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have led to universal MLIPs (uMLIPs) that are pre-trained on diverse datasets, providing opportunities for universal force fields and foundational machine learning models. However, their performance in extrapolating to out-of-distribution complex atomic environments remains unclear. In this study, we highlight a consistent potential energy surface (PES) softening effect in three uMLIPs: M3GNet, CHGNet, and MACE-MP-0, which is characterized by energy and force underprediction in atomic-modeling benchmarks including surfaces, defects, solid-solution energetics, ion migration barriers, phonon vibration modes, and general high-energy states. The PES softening behavior originates primarily from the systematically underpredicted PES curvature, which derives from the biased sampling of near-equilibrium atomic arrangements in uMLIP pre-training datasets. Our findings suggest that a considerable fraction of uMLIP errors are highly systematic, and can therefore be efficiently corrected. We argue for the importance of a comprehensive materials dataset with improved PES sampling for next-generation foundational MLIPs.
- Open Access
- 23 December 2024
- Published by Wiley in Advanced Materials
- Vol. 37 (6) , e2412871
Accelerating the Electrochemical Formation of the δ Phase in Manganese‐Rich Rocksalt Cathodes
Mn‐rich disordered rocksalt materials with Li‐excess (DRX) materials have emerged as a promising class of earth‐abundant and energy‐dense next‐generation cathode materials for lithium‐ion batteries. Recently, an electrochemical transformation to a spinel‐like “δ” phase has been reported in Mn‐rich DRX materials, with improved capacity, rate capability, and cycling stability compared with previous DRX compositions. However, this transformation unfolds slowly over the course of cycling, complicating the development and understanding of these materials. In this work, it is reported that the transformation of Mn‐rich DRX materials to the promising δ phase can be promoted to occur much more rapidly by electrochemical pulsing at elevated temperature, rate, and voltage. To extend this concept, micron‐sized single‐crystal DRX particles are also transformed to the δ phase by the same method, possessing greatly improved cycling stability in the first demonstration of cycling for large, single‐crystal DRX particles. To shed light on the formation and specific structure of the δ phase, X‐ray diffraction, scanning electron nanodiffraction (SEND) and atomic resolution STEM‐HAADF are used to reveal a nanodomain spinel structure with minimal remnant disorder.
- 23 December 2024
- Published by Wiley in Advanced Materials
- Vol. 37 (6) , e2412871
Open AccessAccelerating the Electrochemical Formation of the δ Phase in Manganese‐Rich Rocksalt Cathodes
Mn‐rich disordered rocksalt materials with Li‐excess (DRX) materials have emerged as a promising class of earth‐abundant and energy‐dense next‐generation cathode materials for lithium‐ion batteries. Recently, an electrochemical transformation to a spinel‐like “δ” phase has been reported in Mn‐rich DRX materials, with improved capacity, rate capability, and cycling stability compared with previous DRX compositions. However, this transformation unfolds slowly over the course of cycling, complicating the development and understanding of these materials. In this work, it is reported that the transformation of Mn‐rich DRX materials to the promising δ phase can be promoted to occur much more rapidly by electrochemical pulsing at elevated temperature, rate, and voltage. To extend this concept, micron‐sized single‐crystal DRX particles are also transformed to the δ phase by the same method, possessing greatly improved cycling stability in the first demonstration of cycling for large, single‐crystal DRX particles. To shed light on the formation and specific structure of the δ phase, X‐ray diffraction, scanning electron nanodiffraction (SEND) and atomic resolution STEM‐HAADF are used to reveal a nanodomain spinel structure with minimal remnant disorder.
- 19 December 2024
- Published by American Chemical Society (ACS) in Journal of the American Chemical Society
- Vol. 147 (1) , 223-233
Oxygen Dimerization-Driven Cation Migration Induces Voltage Hysteresis in Disordered Rocksalt Cathodes
Despite the potential to increase the energy limit of Li-rich cathodes by using oxygen redox, its practicality has been limited by the accompanying structural changes and voltage hysteresis. While voltage hysteresis is commonly associated with transition metal (TM) migration and oxygen dimerization, the specific contribution of each is unclear. We provide a mechanistic insight into how each of these changes induces hysteresis in a representative Li-rich disordered rocksalt cathode, Li1.2Mn0.4Ti0.4O2. We reveal that the formation and cleavage of oxygen dimers can occur exceptionally rapidly during the electrochemical process, suggesting that the dimerization process is not directly the cause of voltage hysteresis, contrary to prevailing arguments. Instead, oxygen dimers are found to indirectly exacerbate hysteresis by instigating TM migration, which leads to the evolution of dimer-rich and TM-rich regions within the structure. We demonstrate that TM migration is relatively slower than dimerization and as such contributes to hysteresis by dissipating internal energy during the relaxation of charged electrodes and by inducing cation rearrangement with each cycle.
- 19 December 2024
- Published by American Chemical Society (ACS) in Journal of the American Chemical Society
- Vol. 147 (1) , 223-233
Oxygen Dimerization-Driven Cation Migration Induces Voltage Hysteresis in Disordered Rocksalt Cathodes
Despite the potential to increase the energy limit of Li-rich cathodes by using oxygen redox, its practicality has been limited by the accompanying structural changes and voltage hysteresis. While voltage hysteresis is commonly associated with transition metal (TM) migration and oxygen dimerization, the specific contribution of each is unclear. We provide a mechanistic insight into how each of these changes induces hysteresis in a representative Li-rich disordered rocksalt cathode, Li1.2Mn0.4Ti0.4O2. We reveal that the formation and cleavage of oxygen dimers can occur exceptionally rapidly during the electrochemical process, suggesting that the dimerization process is not directly the cause of voltage hysteresis, contrary to prevailing arguments. Instead, oxygen dimers are found to indirectly exacerbate hysteresis by instigating TM migration, which leads to the evolution of dimer-rich and TM-rich regions within the structure. We demonstrate that TM migration is relatively slower than dimerization and as such contributes to hysteresis by dissipating internal energy during the relaxation of charged electrodes and by inducing cation rearrangement with each cycle.
- Open Access
Overcoming systematic softening in universal machine learning interatomic potentials by fine-tuning
Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have seen the emergence of universal MLIPs (uMLIPs) that are pre-trained on diverse materials datasets, providing opportunities for both ready-to-use universal force fields and robust foundations for downstream machine learning refinements. However, their performance in extrapolating to out-of-distribution complex atomic environments remains unclear. In this study, we highlight a consistent potential energy surface (PES) softening effect in three uMLIPs: M3GNet, CHGNet, and MACE-MP-0, which is characterized by energy and force under-prediction in a series of atomic-modeling benchmarks including surfaces, defects, solid-solution energetics, phonon vibration modes, ion migration barriers, and general high-energy states. We find that the PES softening behavior originates from a systematic underprediction error of the PES curvature, which derives from the biased sampling of near-equilibrium atomic arrangements in uMLIP pre-training datasets. We demonstrate that the PES softening issue can be effectively rectified by fine-tuning with a single additional data point. Our findings suggest that a considerable fraction of uMLIP errors are highly systematic, and can therefore be efficiently corrected. This result rationalizes the data-efficient fine-tuning performance boost commonly observed with foundational MLIPs. We argue for the importance of a comprehensive materials dataset with improved PES sampling for next-generation foundational MLIPs.
Open AccessOvercoming systematic softening in universal machine learning interatomic potentials by fine-tuning
Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have seen the emergence of universal MLIPs (uMLIPs) that are pre-trained on diverse materials datasets, providing opportunities for both ready-to-use universal force fields and robust foundations for downstream machine learning refinements. However, their performance in extrapolating to out-of-distribution complex atomic environments remains unclear. In this study, we highlight a consistent potential energy surface (PES) softening effect in three uMLIPs: M3GNet, CHGNet, and MACE-MP-0, which is characterized by energy and force under-prediction in a series of atomic-modeling benchmarks including surfaces, defects, solid-solution energetics, phonon vibration modes, ion migration barriers, and general high-energy states. We find that the PES softening behavior originates from a systematic underprediction error of the PES curvature, which derives from the biased sampling of near-equilibrium atomic arrangements in uMLIP pre-training datasets. We demonstrate that the PES softening issue can be effectively rectified by fine-tuning with a single additional data point. Our findings suggest that a considerable fraction of uMLIP errors are highly systematic, and can therefore be efficiently corrected. This result rationalizes the data-efficient fine-tuning performance boost commonly observed with foundational MLIPs. We argue for the importance of a comprehensive materials dataset with improved PES sampling for next-generation foundational MLIPs.
- 9 October 2023
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 35 (20) , 8455-8463
Elucidating Structural Transition Dynamics in the Magnesium Cathode MgCr2O4
- 9 October 2023
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 35 (20) , 8455-8463
Elucidating Structural Transition Dynamics in the Magnesium Cathode MgCr2O4
- 11 September 2023
- Published by American Physical Society (APS) in Physical Review Materials
- Vol. 7 (9) , 095801
Impact of the energy landscape on the ionic transport of disordered rocksalt cathodes
Traditional approaches to identify ion-transport pathways often presume an equal probability of occupying all hopping sites and focus entirely on finding the lowest migration barrier channels between them. Although this strategy has been applied successfully to solid-state Li battery materials, which historically have mostly been ordered frameworks, in the emerging class of disordered electrode materials some Li sites can be significantly more stable than others due to a varied distribution of transition metal (TM) environments. Using kinetic Monte Carlo simulations, we show that in such cation-disordered compounds only a fraction of the Li sites connected by the so-called low-barrier “0-TM” channels actually participate in Li diffusion. The Li-diffusion behavior through these sites, which are determined primarily by the voltage applied during Li extraction, can be captured using an effective migration barrier larger than that of the 0-TM barrier itself. The suppressed percolation due to cation disorder can decrease the ionic diffusion coefficient at room temperature by over two orders of magnitude.
- 11 September 2023
- Published by American Physical Society (APS) in Physical Review Materials
- Vol. 7 (9) , 095801
Impact of the energy landscape on the ionic transport of disordered rocksalt cathodes
Traditional approaches to identify ion-transport pathways often presume an equal probability of occupying all hopping sites and focus entirely on finding the lowest migration barrier channels between them. Although this strategy has been applied successfully to solid-state Li battery materials, which historically have mostly been ordered frameworks, in the emerging class of disordered electrode materials some Li sites can be significantly more stable than others due to a varied distribution of transition metal (TM) environments. Using kinetic Monte Carlo simulations, we show that in such cation-disordered compounds only a fraction of the Li sites connected by the so-called low-barrier “0-TM” channels actually participate in Li diffusion. The Li-diffusion behavior through these sites, which are determined primarily by the voltage applied during Li extraction, can be captured using an effective migration barrier larger than that of the 0-TM barrier itself. The suppressed percolation due to cation disorder can decrease the ionic diffusion coefficient at room temperature by over two orders of magnitude.
- Open Access
- 4 June 2023
- Published by Wiley in Advanced Science
- Vol. 10 (23) , e2302086
Opening the Bandgap of Metallic Half‐Heuslers via the Introduction of d–d Orbital Interactions
Half‐Heusler compounds with semiconducting behavior have been developed as high‐performance thermoelectric materials for power generation. Many half‐Heusler compounds also exhibit metallic behavior without a bandgap and thus inferior thermoelectric performance. Here, taking metallic half‐Heusler MgNiSb as an example, a bandgap opening strategy is proposed by introducing the d–d orbital interactions, which enables the opening of the bandgap and the improvement of the thermoelectric performance. The width of the bandgap can be engineered by tuning the strength of the d–d orbital interactions. The conduction type and the carrier density can also be modulated in the Mg1‐xTixNiSb system. Both improved n‐type and p‐type thermoelectric properties are realized, which are much higher than that of the metallic MgNiSb. The proposed bandgap opening strategy can be employed to design and develop new half‐Heusler semiconductors for functional and energy applications.
- 4 June 2023
- Published by Wiley in Advanced Science
- Vol. 10 (23) , e2302086
Open AccessOpening the Bandgap of Metallic Half‐Heuslers via the Introduction of d–d Orbital Interactions
Half‐Heusler compounds with semiconducting behavior have been developed as high‐performance thermoelectric materials for power generation. Many half‐Heusler compounds also exhibit metallic behavior without a bandgap and thus inferior thermoelectric performance. Here, taking metallic half‐Heusler MgNiSb as an example, a bandgap opening strategy is proposed by introducing the d–d orbital interactions, which enables the opening of the bandgap and the improvement of the thermoelectric performance. The width of the bandgap can be engineered by tuning the strength of the d–d orbital interactions. The conduction type and the carrier density can also be modulated in the Mg1‐xTixNiSb system. Both improved n‐type and p‐type thermoelectric properties are realized, which are much higher than that of the metallic MgNiSb. The proposed bandgap opening strategy can be employed to design and develop new half‐Heusler semiconductors for functional and energy applications.
- 8 August 2022
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry A
- Vol. 10 (36) , 18737-18744
Discovery of triple half-Heusler Mg 2 VNi 3 Sb 3 with low thermal conductivity
Triple half-Heusler Mg 2 VNi 3 Sb 3 was successfully synthesized by following an unconventional valence balance strategy. A new strategy to explore the huge compositional space for extended tunability of intermetallic compounds was demonstrated.
- 8 August 2022
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry A
- Vol. 10 (36) , 18737-18744
Discovery of triple half-Heusler Mg 2 VNi 3 Sb 3 with low thermal conductivity
Triple half-Heusler Mg 2 VNi 3 Sb 3 was successfully synthesized by following an unconventional valence balance strategy. A new strategy to explore the huge compositional space for extended tunability of intermetallic compounds was demonstrated.
- 23 June 2022
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 4 (7) , 3392-3398
Structural Understanding of the Slater–Pauling Electron Count in Defective Heusler Thermoelectric TiFe1.5Sb as a Valence Balanced Semiconductor
- 23 June 2022
- Published by American Chemical Society (ACS) in ACS Applied Electronic Materials
- Vol. 4 (7) , 3392-3398
Structural Understanding of the Slater–Pauling Electron Count in Defective Heusler Thermoelectric TiFe1.5Sb as a Valence Balanced Semiconductor