Session A60: Computational Design, Understanding and Discovery of Novel Materials I

High-Entropy Oxides: Multifunctionality Enabled through Chemical Disorder

Entropy engineering offers access to a vast, unexplored materials discovery space by utilizing configurational entropy as a thermodynamic driver towards new multicomponent crystalline solids. Our interdisciplinary team within the Materials Research Science and Engineering Center (MRSEC) at Penn State explores high-entropy oxide (HEO) materials in which the cation sublattices are occupied by many elements at random. The ability to stabilize elements in unusual states within the high-entropy matrix offers an attractive opportunity for functional property engineering in complex oxides, specifically ionic and transparent conductors, relaxor ferroelectrics, and strongly correlated materials. Our interdisciplinary team utilizes first-principles computational predictions to accelerate the discovery of novel HEO compositions as well as for tuning and understanding of functional properties. The combination of recent improved functionals as well as uniquely defined descriptors allow us to traverse the complex HEO composition space in a tractable manner. Our team's efforts hope to push the materials discovery frontier beyond the classical realm of enthalpic stability by utilizing entropy for the next generation of complex-oxide materials.   

High-entropy alloy and intermetallic design by Super-Cell Random Approximates (SCRAPs)

Structural models of high-entropy alloys with specific short-range order were developed through a highly accelerated hybrid Cuckoo Search – an evolutionary algorithm that combines Levi flights for global searches with Monte Carlo for local searches [Nat. Comput. Sci. 1, 54 (2021)].  We extend SCRAPs (ver. 2.0) to rapidly create specific (h k l)-oriented bulk, interface, or surface models by utilizing an efficient algorithm by Hermann [Surf. Rev. Lett. 4, 1063 (1997)] as the starting unit. As an example, we find an optimal solution from 10⁷³ unique configurations in ~13 seconds that also exhibits strong scaling when run on parallel architectures. These SCRAPs 2.0 representative structural solutions can then be used within density-functional theory to design properties of high-entropy alloy and intermetallic surfaces, such as to design catalytic surfaces for hydrogen-evolution reactions. Finally, we demonstrate that the integration of an Integer Linear Programming Library (LPSolve) adds functionality and improves efficiency by avoiding searches when a solution in unfeasible and setting system-specific variability to prevent Monte Carlo from stagnating.   

Mixed Enthalpy-Entropy Descriptor Guided Discovery of 2D High-Entropy Chalcogenides for Energy Applications

Over the past few years, the discovery of high-entropy materials (HEMs) has sparked significant interest in condensed matter physics and materials science due to their potential for a wide range of applications. These materials have been of growing interest, but the design of HEMs faces daunting challenges, which have been recently overcome by the Mixed Enthalpy-Entropy Descriptor (MEED) that successfully predicted several experimentally reported HEMs from first principles. In this work, the MEED has been used to screen the materials space of 2D chalcogenides of metals (Hf, Nb, Mo, Ta, Ti, V, W, and Zr) in the 2H, 1T, and 1T' phases, comprising four, five, and six principal metal elements. Interestingly, MEED captures the structural aspects relevant to the synthesis of HEMs in their respective polymorphs, further showcasing its unique capability. Additionally, based on MEED predictions for 2D high-entropy chalcogenides, several top-candidate high-entropy tellurides have been successfully synthesized. Furthermore, the results indicate that these high-entropy tellurides, which can blend the combinatorial properties of HEMs with the large surface area of 2D materials, exhibit exceptional characteristics for applications in batteries and catalysis.   

Discovery of new clathrates by computational screening, and directed synthesis

Charge-balanced inorganic clathrates are semiconducting materials with unique properties, such as ultra-low thermal conductivity, superconductivity, and high-density ion storage. They consist of covalent frames of nanometer-size polyhedral cages encapsulating metallic guest atoms. While the frame of conventional clathrates mostly consists of group IV elements (Si, Ge, Sn), we have recently discovered new stable intermetallic clathrates with III-V frames and enticing transport properties.

In this work, we combine Zintl compositional predictions, high-throughput density functional theory calculations, and machine learning to screen several thousands of compositions of III-V and II-VI type-I clathrate frames, mixed with transition metals, to identify stability trends and families of potentially stable compounds. In particular, a novel workflow is developed, which uses transfer learning to refine convolutional graph neural networks for predicting the site ordering of clathrate superstructures, as well as their equilibrium density and decomposition energy.

Theoretical predictions are tested by ex-situ synthesis and crystal growth, which have revealed new clathrates and clathrate-like compounds. Single-phase synthesis and first-principles calculations enable the joint theoretical and experimental exploration of their transport properties, revealing promising properties for thermoelectric energy conversion.   

Martensitic Phase Transitions in Complex NiTi-Based Shape Memory Alloys from First-Principles Calculations

Recent rapid progresses in physics theory and computational power have made it possible to accurately predict the phase transitions and martensitic transition temperatures (MTTs) in shape memory alloys (SMAs) from first principles. However, previously theory and calculations were applied only to study highly ordered stoichiometric binary alloys such as NiTi, PdTi and NiHf. Here we report on our recent first-principles investigations on Ni_0.5Ti_0.5-xHf_x and Pd_xNi_0.5-xTi_0.5 ternaries and off-stoichiometric NiTi. We show that the predicted martensitic phase transitions in these complex SMAs are in good agreement with experimental findings. In particular, the calculated MTTs for all these compositions are within ~ 100 K compared with the corresponding measured data, and our results also reveal the physical origin of the striking asymmetry in MTT of the off-stoichiometric NiTi near equiatomic compositions. We will address various techniques developed to overcome the difficulty encountered in studying ternaries and off-stoichiometric binaries associated with disorder and/or much lowered symmetry. Our theoretical approach is expected to be a broadly applicable and predictive theory for designing complex SMAs with desirable properties.   

Redox properties of transition metal oxides NiO2 and CoO2 from a defect perspective

The splitting of water is considered the cleanest method for harvesting renewable energy, offering both environmental and economic benefits. Here, I will present our study on the redox properties of two layered transition metal oxides NiO₂ and CoO₂. There is a consensus that an open d-shell transition-metal atom in a specific oxidation state is desirable for favorable catalysis. The oxidation state of transition-metal can be tuned  by using  alkali metals, thereby restructuring the electronic properties of these layered materials to make them favorable for catalysis. Our study employs the density functional HSE06 and r2SCAN+rVV10+U to search for the polaron that can produce the favorable e_g¹ configuration.   

Machine learning based computational methods to analyze structural characterization in soft materials

Soft materials researcher aiming to establish design-structure-property relationships have to rely on structural characterization from multiple complementary techniques to obtain a holistic understanding of structures within their newly designed material. Depending on the availability and accessibility of the different characterization techniques (e.g., scattering, microscopy, spectroscopy), each research facility or academic research lab may have access to high-throughput capability in one technique but face limitations (sample preparation, resolution, access time) with other technique(s). Furthermore, one type of structural characterization data may be easier to interpret than another (e.g., microscopy images are easier to interpret than small angle scattering profiles). Thus, it is useful to have machine learning models that can be trained on paired structural characterization data from multiple techniques (easy and difficult to interpret, fast and slow in data collection or sample preparation), so that the model can generate one set of characterization data from the other. In this talk I will discuss one such machine learning model called PairVAE that pairs small angle scattering results (i.e., information about bulk morphology) and electron microscopy images (i.e., information about two-dimensional local structure). Using paired SAXS and SEM data of newly observed block copolymer assembled morphologies [open access data from Doerk G.S., et al. Science Advances. 2023 Jan 13;9(2): eadd3687], we trained the PairVAE model.  After successful training, PairVAE is able to generate SEM images of the block copolymer morphology when it takes as input that sample’s corresponding SAXS 2D pattern, and vice versa.   

High-throughput virtual screening for high thermal conductivity polymer crystals using first-principles calculation

Polymer crystals with high thermal conductivity are efficient and cost-effective materials for a range of uses, including heat exchangers and additive manufacturing. In this work, a data-driven approach for finding the high thermal conductivity polymer crystals has been proposed. Moreover, first-principles phonon calculations were used to assess the lattice thermal conductivity (LTC), phonon lifetimes, and modal heat capacities of polymer crystals. A set of 1073 polymer crystal structures obtained from the polymer genome datasets was employed for a search of high thermal conductivity polymer crystals. Moreover, an optimized structural descriptor correlating bulk modulus curvature and LTC was used for high-throughput computational screening. In this context, we obtained Polymethylenimine (PMI), Poly (methylene oxide) (PMO), and Polyamide (PA) polymers with high thermal conductivities of 21.81, 94.95, and 65.27 W/m·K, respectively, at 300K. In addition, the lattice thermal conductivity of PMO exceeded 100 W/m·K in the temperature range of 100 to 270K. In conclusion, the integration of first-principles calculations with a data-driven approach offers a novel strategy to expedite the discovery of high thermal conductivity polymer crystals.   

Reproducible workflows for verification and optimization of solid-state pseudopotentials

The quality of pseudopotentials (PPs), a key approximation for planewaves implementations of density-functional-theory (DFT), is crucial in obtaining trustworthy predictions of material properties. Nevertheless, verifying PPs is not trivial: it requires significant expertise in running DFT simulations and can involve hundreds of DFT runs. Optimization is even harder: every iteration requires this verification step, further increasing the number of calculations needed. We address this issue by developing a toolkit based on the AiiDA [1] workflow engine and the AiiDAlab [2] interface to automate and streamline the verification and optimization of PPs. Using the toolkit, two curated PP libraries tagged with “efficiency” and “precision” are improved, further extending the SSSP library [3]: “efficiency” is targeted at high-throughput screening, and “precision” at high-accuracy modeling. This effort has been facilitated by the development of AiiDA SSSP workflows that ensure full reproducibility. The toolkit is user-friendly and can be used to verify and optimize PPs for more specific usages, e.g., core-hole PPs needed in XPS simulations.

[1] S. P. Huber et al., Sci. Data 7, 300 (2020).

[2] A. V. Yakutovich et al., Comput. Mater. Sci. 188, 110165 (2021).

[3] G. Prandini et al., npj Comput. Mater. 4, 72 (2018).   

Thermodynamic and Electronic Properties of Semiconducting High-Entropy Halide Perovskites Materials through First-Principles Calculations

High-entropy materials are known for forming single phase solid solution and having the potential for property designs. They have emerged as promising candidates for sustainable energy technologies. However, their high synthesis temperatures present challenges. In contrast, halide-based perovskites offer a low-temperature synthesis alternative, providing a prospective way to bring this excellent-performance material family into real-life manufacturing and usage. 

 

In this work, we employ density functional theory (DFT) calculations to investigate the formation enthalpy and electronic band structure of high-entropy halide perovskites designed on the cubic vacancy-ordered double-perovskite structure Cs₂MCl₆ (M=Zr⁴⁺, Sn⁴⁺, Te⁴⁺, Hf⁴⁺, Re⁴⁺, Os⁴⁺, Ir⁴⁺ or Pt⁴⁺). Our goal is to provide atomistic understanding of their electronic properties and thermodynamic stability. We analyzed the eight constituent compounds, unveiling the fundamental physical mechanisms governing transport properties which further help understand the behavior of the high-entropy phases. We assessed the thermodynamic stability of the high-entropy phases by comparing the enthalpy and entropy of mixing to their constituent phases. By leveraging high-throughput calculations and interpolation calculations, our study maps the properties of the high-entropy phases across a broad composition range. Our multidisciplinary approach provide accurate insights into high-entropy perovskites that can be synthesized at low temperatures, with the potential to advance clean and affordable energy technologies.