Machine Learning-Enabled Superior Energy Storage in Ferroelectric Films with a Slush-Like Polar State.

Heterogeneities in structure and polarization have been employed to enhance the energy storage properties of ferroelectric films. The presence of nonpolar phases, however, weakens the net polarization. Here, we achieve a slush-like polar state with fine domains of different ferroelectric polar phases by narrowing the large combinatorial space of likely candidates using machine learning methods. The formation of the slush-like polar state at the nanoscale in cation-doped BaTiO3 films is simulated by phase field simulation and confirmed by aberration-corrected scanning transmission electron microscopy. The large polarization and the delayed polarization saturation lead to greatly enhanced energy density of 80 J/cm3 and transfer efficiency of 85% over a wide temperature range. Such a data-driven design recipe for a slush-like polar state is generally applicable to quickly optimize functionalities of ferroelectric materials.

Accelerated search for BaTiO3-based piezoelectrics with vertical morphotropic phase boundary using Bayesian learning.

An outstanding challenge in the nascent field of materials informatics is to incorporate materials knowledge in a robust Bayesian approach to guide the discovery of new materials. Utilizing inputs from known phase diagrams, features or material descriptors that are known to affect the ferroelectric response, and Landau-Devonshire theory, we demonstrate our approach for BaTiO3-based piezoelectrics with the desired target of a vertical morphotropic phase boundary. We predict, synthesize, and characterize a solid solution, (Ba0.5Ca0.5)TiO3-Ba(Ti0.7Zr0.3)O3, with piezoelectric properties that show better temperature reliability than other BaTiO3-based piezoelectrics in our initial training data.

Low-Field-Triggered Large Magnetostriction in Iron-Palladium Strain Glass Alloys.

Development of miniaturized magnetostriction-associated devices requires low-field-triggered large magnetostriction. In this study, we acquired a large magnetostriction (800 ppm) triggered by a low saturation field (0.8 kOe) in iron-palladium (Fe-Pd) alloys. Magnetostriction enhancement jumping from 340 to 800 ppm was obtained with a slight increase in Pd concentration from 31.3 to 32.3 at. %. Further analysis showed that such a slight increase led to suppression of the long-range ordered martensitic phase and resulted in a frozen short-range ordered strain glass state. This strain glass state possessed a two-phase nanostructure with nanosized frozen strain domains embedded in the austenite matrix, which was responsible for the unique magnetostriction behavior. Our study provides a way to design novel magnetostrictive materials with low-field-triggered large magnetostriction.

Direct evidence for local symmetry breaking during a strain glass transition.

Strain glass transition is a unique nanoscale displacive transition with local symmetry breaking while maintaining the macroscopic symmetry or average structure unchanged. It usually occurs in the "nonmartensitic" composition range of a martensitic system. So far, only indirect evidence exists for such a transition, essentially from macroscopic measurements and low-resolution transmission electron microscopy observations, and there is a lack of direct evidence for the speculated local symmetry breaking and the sluggish nature of the glass transition. In this Letter we report in situ high-resolution transmission electron microscopy observations on a Ti50(Pd41Cr9) strain glass alloy and direct evidence for these key issues. Our results show that at temperatures well above the strain glass transition temperature (Tg), the lattice is essentially an undistorted B2 structure. With approaching Tg, the local symmetry breaking gradually occurs with the formation and growth of nanomartensite clusters with a combined stacking period of three and four plane intervals, but the average structure measured by x-ray diffraction remains B2. These nanomartensite clusters become finally frozen below Tg. Our results provide not only a microscopic basis for the macroscopic properties of strain glass, but also new insights into a range of possible applications of this unique class of materials.

Noise-Aware Active Learning to Develop High-Temperature Shape Memory Alloys with Large Latent Heat.

Shape memory alloys (SMAs) with large latent heat absorbed/released during phase transformation at elevated temperatures benefit their potential application on thermal energy storage (TES) in high temperature environment like power plants, etc. The desired alloys can be designed quickly by searching the vast component space of doped NiTi-based SMAs via data-driven method, while be challenging with the noisy experimental data. A noise-aware active learning strategy is proposed to accelerate the design of SMAs with large latent heat at elevated phase transformation temperatures based on noisy data. The optimal noise level is estimated by minimizing the model error with incorporation of a range of noise levels as noise hyper-parameters into the noise-aware Kriging model. The employment of this strategy leads to the discovery of the alloy with latent heat of -36.08 J g-1, 9.2% larger than the best value (-33.04 J g-1) in the original training dataset within another four experiments. Additionally, the alloy represents high austenite finish temperature (481.71°C) and relatively small hysteresis. This promotes the latent heat TES application of SMAs in high temperature circumstance. It is expected that the noise-aware approach can be convenient for the accelerated materials design via the data-driven method with noisy data.

Ultrahigh Elastic Energy Storage in Nanocrystalline Alloys with Martensite Nanodomains.

Elastic materials that store and release elastic energy play pivotal roles in both macro and micro mechanical systems. Uniting high elastic energy density and efficiency is crucial for emerging technologies such as artificial muscles, hopping robots, and unmanned aerial vehicle catapults, yet it remains a significant challenge. Here, a nanocrystalline structure embedded with elliptical martensite nanodomains in ferroelastic alloys was utilized to enable high yield strength, large recoverable strain, and low energy dissipation simultaneously. As a result, the designed Ti-Ni-V alloys demonstrate ultrahigh energy density (>40 MJ m-3) with ultrahigh efficiency (>93%) and exceptional durability. This concept, which combines nano-sized embryos to minimize energy dissipation of psuedo-elasticity and employs a fine-grained structure to enhance yield strength, can be applied to other ferroelastic materials. Furthermore, it holds promise for the development of phase transformation-involved functionalities such as high-performance dielectric energy storage, ultralow-hysteresis magnetostrain, and high-efficiency solid-state caloric cooling.

Efficient roller-driven elastocaloric refrigerator.

Elastocaloric cooling has experienced fast development over the past decade owing to its potential to reshape the refrigeration industry. While the solid-state elastocaloric refrigerant is emission-free, the efficiency of the state-of-the-art elastocaloric cooling systems is not sufficient yet to reduce carbon emissions during operation. In this study, we double the coefficient of performance, the most commonly used efficiency metric, via the synergy of material-level advances in TiNiCu and the system-level roller-driven mechanism capable of recovering kinetic energy. On the materials level, a 125% improvement in coefficient of performance is illustrated in TiNiCu compared to NiTi, empowered by the B2-B19 martensitic transformation with improved lattice compatibility and the grain boundary strengthening from the nanocrystalline structure. On the system level, owing to the properly sized angular momentum in rotating parts, 78% work recovery efficiency is reported, transcending the theoretical limit previously unattainable without kinetic energy recovery. This confluence of materials and mechanical innovations propels elastocaloric cooling systems into a new realm of efficiency and paves the way for their practical application.

Enhanced Piezoelectric Response at Nanoscale Vortex Structures in Ferroelectrics.

The piezoelectric response is a measure of the sensitivity of a material's polarization to stress or its strain to an applied field. Using in operando X-ray Bragg coherent diffraction imaging, we observe that topological vortices are the source of a 5-fold enhancement of the piezoelectric response near the vortex core. The vortices form where several low-symmetry ferroelectric phases and phase boundaries coalesce. Unlike bulk ferroelectric solid solutions in which a large piezoelectric response is associated with coexisting phases in the proximity of the triple point, the largest responses for pure BaTiO3 at the nanoscale are in spatial regions of extremely small spontaneous polarization at vortex cores. The response decays inversely with polarization away from the vortex, analogous to the behavior in bulk ceramics as the cation compositions are varied away from the triple point. We use first-principles-based molecular dynamics to augment our observations, and our results suggest that nanoscale piezoelectric materials with a large piezoelectric response can be designed within a parameter space governed by vortex cores. Our findings have implications for the development of next-generation nanoscale piezoelectric materials.

Electric Field Switching of Magnon Spin Current in a Compensated Ferrimagnet.

Manipulation of directional magnon propagation, known as magnon spin current, is essential for developing magnonic devices featuring nonvolatile functionalities and ultralow power consumption. Magnon spin current can usually be modulated by magnetic field or current-induced spin torques. However, these approaches may lead to energy dissipation due to Joule heating. Electric-field switching of magnon spin current without charge current is highly preferred but challenging to realize. By integrating magnonic and piezoelectric materials, the manipulation of the magnon spin current generated by the spin Seebeck effect in the ferrimagnetic insulator Gd3Fe5O12 (GdIG) film on a piezoelectric substrate is demonstrated. Reversible electric-field switching of magnon polarization without applied charge current is observed. Through strain-mediated magnetoelectric coupling, the electric field induces the magnetic compensation transition between two magnetic states of the GdIG, resulting in its magnetization reversal and the simultaneous switching of magnon spin current. This work establishes a prototype material platform that paves the way for developing magnon logic devices characterized by all electric field reading and writing and reveals the underlying physics principles of their functions.

Erratum: Upper limit of the transition temperature of superconducting materials.

[This corrects the article DOI: 10.1016/j.patter.2022.100609.].

Upper limit of the transition temperature of superconducting materials.

Why are the transition temperatures (T c) of superconducting materials so different? The answer to this question is not only of great significance in revealing the mechanism of high-T c superconductivity but also can be used as a guide for the design of new superconductors. However, so far, it is still challenging to identify the governing factors affecting the T c. In this work, with the aid of machine learning and first-principles calculations, we found a close relevance between the upper limit of the T c and the energy-level distribution of valence electrons. It implies that some additional inter-orbital electron-electron interaction should be considered in the interpretation of high-T c superconductivity.

Machine-Learning-Enabled Prediction of Adiabatic Temperature Change in Lead-Free BaTiO3-Based Electrocaloric Ceramics.

In this paper, we develop a data-driven machine learning (ML) approach to predict the adiabatic temperature change (ΔT) in BaTiO3-based ceramics as a function of chemical composition, temperature, and applied electric field. The data set was curated from a survey of published electrocaloric measurements. Each chemical composition was represented by elemental descriptors of A-site and B-site elements. Pair-wise statistical correlation analysis was used to remove linearly correlated descriptors. We trained two separate regression-based ML models for indirect and direct measurements and found that both are capable of capturing the general trend of the temperature vs ΔT curve for various applied electric fields. We then complemented the regression models with a classification learning model that predicts the expected phase as a function of chemical composition and temperature. The combined regression and classification learning ML models predict a global maxima in ΔT near rhombohedral to cubic or tetragonal to cubic phase transition regions. An interactive, open source web application is developed to enable interested users to query our trained models and accelerate the design of novel BaTiO3-based ceramics with targeted phase and ΔT properties for electrocaloric applications.

Boson-peak-like anomaly caused by transverse phonon softening in strain glass.

Strain glass is a glassy state with frozen ferroelastic/martensitic nanodomains in shape memory alloys, yet its nature remains unclear. Here, we report a glassy feature in strain glass that was thought to be only present in structural glasses. An abnormal hump is observed in strain glass around 10 K upon normalizing the specific heat by cubed temperature, similar to the boson peak in metallic glass. The simulation studies show that this boson-peak-like anomaly is caused by the phonon softening of the non-transforming matrix surrounding martensitic domains, which occurs in a transverse acoustic branch not associated with the martensitic transformation displacements. Therefore, this anomaly neither is a relic of van Hove singularity nor can be explained by other theories relying on structural disorder, while it verifies a recent theoretical model without any assumptions of disorder. This work might provide fresh insights in understanding the nature of glassy states and associated vibrational properties.

Knowledge-Based Descriptor for the Compositional Dependence of the Phase Transition in BaTiO3-Based Ferroelectrics.

Descriptors play a central role in constructing composition-structure-property relationships to guide materials design. We propose a material descriptor, δτ, for the composition dependence of the Curie temperature (Tc) on single doping elements in BaTiO3 ferroelectrics, which is then generalized to a linear combination of multiple dopants in the solid solutions. The descriptor δτ depends linearly on the Curie temperature and also serves to separate the ferroelectric phase from the relaxor phase. We compare δτ to other commonly used descriptors such as the tolerance factor, electronegativity, and ionic displacement. By using regression analysis on our assembled experimental data, we show how it outperforms other descriptors. We use the trained machine-learned models to predict compositions in our search space with the largest ferroelectric, dielectric, and piezoelectric properties, namely, d33, electrostrain, and recoverable energy storage density. We experimentally verify our predictions for Tc and classification into ferroelectrics and relaxors by synthesizing and characterizing six solid solutions in BaTiO3 ferroelectrics. Our definition of δτ can shed light on the design of knowledge-based descriptors in other systems such as Pb-based and Bi-based solid solutions.

Determining Multi-Component Phase Diagrams with Desired Characteristics Using Active Learning.

Herein, we demonstrate how to predict and experimentally validate phase diagrams for multi-component systems from a high-dimensional virtual space of all possible phase diagrams involving several elements based on small existing experimental data. The experimental data for bulk phases for known systems represents a sampling from this space, and screening the space allows multi-component phase diagrams with given design criteria to be built. This approach uses machine learning methods to predict phase diagrams and Bayesian experimental design to minimize experiments for refinement and validation, all within an active learning loop. The approach is proven by predicting and synthesizing the ferroelectric ceramic system (1-ω)(Ba0.61Ca0.28Sr0.11TiO3)-ω(BaTi0.888Zr0.0616Sn0.0028Hf0.0476O3) with a relatively high transition temperature and triple point, as well as the NiTi-based pseudo-binary phase diagram (1-ω)(Ti0.309Ni0.485Hf0.20Zr0.006)-ω(Ti0.309Ni0.485Hf0.07Zr0.068Nb0.068) designed for high transition temperature (ω ⩽ 1). Each phase diagram is validated and optimized through only three new experiments. The complexity of these compounds is beyond the reach of today's computational methods.

Machine Learning Magnetic Parameters from Spin Configurations.

Hamiltonian parameters estimation is crucial in condensed matter physics, but is time- and cost-consuming. High-resolution images provide detailed information of underlying physics, but extracting Hamiltonian parameters from them is difficult due to the huge Hilbert space. Here, a protocol for Hamiltonian parameters estimation from images based on a machine learning (ML) architecture is provided. It consists in learning a mapping between spin configurations and Hamiltonian parameters from a small amount of simulated images, applying the trained ML model to a single unexplored experimental image to estimate its key parameters, and predicting the corresponding materials properties by a physical model. The efficiency of the approach is demonstrated by reproducing the same spin configuration as the experimental one and predicting the coercive field, the saturation field, and even the volume of the experiment specimen accurately. The proposed approach paves a way to achieve a stable and efficient parameters estimation.

The Search for BaTiO3-Based Piezoelectrics With Large Piezoelectric Coefficient Using Machine Learning.

We employ a data-driven approach to search for BaTiO3-based piezoelectrics with large piezoelectric coefficient d33. Our approach uses a surrogate model to make predictions of d33 with uncertainties, followed by a design step that selects the next optimal compound to synthesize. We compare several combinations of choices of the model and design selection strategies on the training data assembled from many experiments that we have previously performed, and we choose the best two performers for guiding new experiments. This adaptive design strategy is iterated five times and in each iteration, four new compounds are synthesized based on the two different design selection criteria. The best new compound found in this work is (Ba0.85Ca0.15)(Ti0.91Zr0.09)O3 with a d33 of 362 pC/N, compared to the best compound BCT-0.5BZT in the training data with a d33 of ~610 pC/N. Our conclusion from this study is that although our model describes well most of the available d33 data, the especially large value for BCT-0.5BZT is difficult to fit with any surrogate model and emphasizes the need to combine a physics-based approach with a pure data-driven approach used in this study.

Accelerated Search for BaTiO3-Based Ceramics with Large Energy Storage at Low Fields Using Machine Learning and Experimental Design.

The problem that is considered is that of maximizing the energy storage density of Pb-free BaTiO3-based dielectrics at low electric fields. It is demonstrated that how varying the size of the combinatorial search space influences the efficiency of material discovery by comparing the performance of two machine learning based approaches where different levels of physical insights are involved. It is started with physics intuition to provide guiding principles to find better performers lying in the crossover region in the composition-temperature phase diagram between the ferroelectric phase and relaxor ferroelectric phase. Such an approach is limiting for multidopant solid solutions and motivates the use of two data-driven machine learning and design strategies with a feedback loop to experiments. Strategy I considers learning and property prediction on all the compounds, and strategy II learns to preselect compounds in the crossover region on which prediction is carried out. By performing only two active learning loops via strategy II, the compound (Ba0.86Ca0.14)(Ti0.79Zr0.11Hf0.10)O3 is synthesized with the largest energy storage density ≈73 mJ cm-3 at a field of 20 kV cm-1, and an insight into the relative performance of the strategies using varying levels of knowledge is provided.

Enhanced magnetism in lightly doped manganite heterostructures: strain or stoichiometry?

Lattice mismatch induced epitaxial strain has been widely used to tune functional properties in complex oxide heterostructures. Apart from the epitaxial strain, a large lattice mismatch also produces other effects including modulations in microstructure and stoichiometry. However, it is challenging to distinguish the impact of these effects from the strain contribution to thin film properties. Here, we use La0.9Sr0.1MnO3 (LSMO), a lightly doped manganite close to the vertical phase boundary, as a model system to demonstrate that both epitaxial strain and cation stoichiometry induced by strain relaxation contribute to functionality tuning. The thinner LSMO films are metallic with a greatly enhanced TC which is 97 K higher than the bulk value. Such anomalies in TC and transport cannot be fully explained by the epitaxial strain alone. Detailed microstructure analysis indicates La deficiency in thinner films and twin domain formation in thicker films. Our results have revealed that both epitaxial strain and strain relaxation induced stoichiometry/microstructure modulations contribute to the modified functional properties in lightly doped manganite perovskite thin films.

Multi-objective Optimization for Materials Discovery via Adaptive Design.

Guiding experiments to find materials with targeted properties is a crucial aspect of materials discovery and design, and typically multiple properties, which often compete, are involved. In the case of two properties, new compounds are sought that will provide improvement to existing data points lying on the Pareto front (PF) in as few experiments or calculations as possible. Here we address this problem by using the concept and methods of optimal learning to determine their suitability and performance on three materials data sets; an experimental data set of over 100 shape memory alloys, a data set of 223 M2AX phases obtained from density functional theory calculations, and a computational data set of 704 piezoelectric compounds. We show that the Maximin and Centroid design strategies, based on value of information criteria, are more efficient in determining points on the PF from the data than random selection, pure exploitation of the surrogate model prediction or pure exploration by maximum uncertainty from the learning model. Although the datasets varied in size and source, the Maximin algorithm showed superior performance across all the data sets, particularly when the accuracy of the machine learning model fits were not high, emphasizing that the design appears to be quite forgiving of relatively poor surrogate models.