Knowledge-driven learning, optimization, and experimental design under uncertainty for materials discovery

Significant acceleration of the future discovery of novel functional materials requires a fundamental shift from the current materials discovery practice, which is heavily dependent on trial-and-error campaigns and high-throughput screening, to one that builds on knowledge-driven advanced informatics techniques enabled by the latest advances in signal processing and machine learning. In this review, we discuss the major research issues that need to be addressed to expedite this transformation along with the salient challenges involved. We especially focus on Bayesian signal processing and machine learning schemes that are uncertainty aware and physics informed for knowledge-driven learning, robust optimization, and efficient objective-driven experimental design.

Deep learning-accelerated computational framework based on Physics Informed Neural Network for the solution of linear elasticity

The paper presents an efficient and robust data-driven deep learning (DL) computational framework developed for linear continuum elasticity problems. The methodology is based on the fundamentals of the Physics Informed Neural Networks (PINNs). For an accurate representation of the field variables, a multi-objective loss function is proposed. It consists of terms corresponding to the residual of the governing partial differential equations (PDE), constitutive relations derived from the governing physics, various boundary conditions, and data-driven physical knowledge fitting terms across randomly selected collocation points in the problem domain. To this end, multiple densely connected independent artificial neural networks (ANNs), each approximating a field variable, are trained to obtain accurate solutions. Several benchmark problems including the Airy solution to elasticity and the Kirchhoff-Love plate problem are solved. Performance in terms of accuracy and robustness illustrates the superiority of the current framework showing excellent agreement with analytical solutions. The present work combines the benefits of the classical methods depending on the physical information available in analytical relations with the superior capabilities of the DL techniques in the data-driven construction of lightweight, yet accurate and robust neural networks. The models developed herein can significantly boost computational speed using minimal network parameters with easy adaptability in different computational platforms.

High throughput exploration of the oxidation landscape in high entropy alloys

High entropy alloys (HEAs) have gained interest for structural applications in extreme environments. With a potentially vast chemical and phase space, there are significant opportunities to discover superior performing alloys. Crucial for most high-temperature applications is understanding and mitigating the oxidation behavior of these chemically complex alloys. Most experimental and computational HEA studies have focused on a limited set of compositions and only a fraction of these compositions have been characterized for oxidation. We present a high-throughput framework that utilizes density-functional theory (DFT) in concert with a combined machine-learning model and grand-canonical linear programming for assessing phase stability, phase-fraction, chemical activity and high-temperature survivability of arbitrary HEAs. This framework considers temperature dependent contributions to the Gibbs energy of the competing phases arising from short-range order and vibrational entropy. We demonstrate the effectiveness of the framework by assessing the thermodynamic stability, oxidation behavior, chemical activity, and phase decomposition of body-centered cubic Mo-W-Ta-Ti-Zr refractory HEAs. A total of 51 compositions were analyzed and ranked in order of their survivability based on the Pareto-front analysis. Oxidation was performed at 1373 K on four samples in air showing the difference in oxidation behavior determined experimentally through scale thickness and their mass changes. The insights on oxidation behavior presented in this work will enable the fast assessment of technologically useful HEAs needed for future structural application in extreme conditions.

Navigating the design space of inorganic materials synthesis using statistical methods and machine learning

Data-driven approaches have brought about a revolution in manufacturing; however, challenges persist in their applications to synthetic strategies. Their application to the deterministic navigation of reaction trajectories to stabilize crystalline solids with precise composition, atomic connectivity, microstructural dimensionality, and surface structure remains a frontier in inorganic materials research. The design of synthetic methodologies for the preparation of inorganic materials is often inefficient in terms of exploration of potentially vast design spaces spanning multiple process variables, reaction sequences, as well as structural parameters and reactivities of precursors and structure-directing agents. Reported synthetic methods are further limited in terms of the insight they provide into underlying chemical and physical principles. The recent surge in interest in accelerating the discovery of new materials can be considered as an opportunity to re-evaluate our approach to materials synthesis, and for considering new frameworks for exploration that are systematic and strategic in approach. Herein, we outline with the help of several illustrative examples, the challenges, opportunities, and limitations of data-driven synthesis design. The account collates discussion of design-of-experiments sampling methods, machine learning modeling, and active learning to develop experimental workflows that accelerate the experimental navigation of synthetic landscapes.

Enhancement of the selectivity of MXenes (M2C, M = Ti, V, Nb, Mo) via oxygen-functionalization: promising materials for gas-sensing and -separation

Mo₂CO₂ and V₂CO₂ reveal very good selectivity toward NO, while Nb₂CO₂ and Ti₂CO₂ show very good selectivity toward NH₃.

On the stochastic phase stability of Ti2AlC-Cr2AlC

The quest towards expansion of the M _n+1AX _n design space has been accelerated with the recent discovery of several solid solution and ordered phases involving at least two M _n+1AX _n end members. Going beyond the nominal M _n+1AX _n compounds enables not only fine tuning of existing properties but also entirely new functionality. This search, however, has been mostly done through painstaking experiments as knowledge of the phase stability of the relevant systems is rather scarce. In this work, we report the first attempt to evaluate the finite-temperature pseudo-binary phase diagram of the Ti₂AlC-Cr₂AlC via first-principles-guided Bayesian CALPHAD framework that accounts for uncertainties not only in ab initio calculations and thermodynamic models but also in synthesis conditions in reported experiments. The phase stability analyses are shown to have good agreement with previous experiments. The work points towards a promising way of investigating phase stability in other MAX Phase systems providing the knowledge necessary to elucidate possible synthesis routes for M _n+1AX _n systems with unprecedented properties.

The complete reaction mechanism of H2S desulfurization on an anatase TiO2 (001) surface: a density functional theory investigation

An anatase TiO₂ (001) surface is active and selective toward water production and results in the modification of the surface by forming S-doped TiO₂, which enhances its photocatalytic activity.

Correction: Phase stability in nanoscale material systems: extension from bulk phase diagrams

Correction for 'Phase stability in nanoscale material systems: extension from bulk phase diagrams' by Saurabh Bajaj et al., Nanoscale, 2015, 7, 9868-9877.

Phase stability in nanoscale material systems: extension from bulk phase diagrams

Phase diagrams of multi-component systems are critical for the development and engineering of material alloys for all technological applications. At nano dimensions, surfaces (and interfaces) play a significant role in changing equilibrium thermodynamics and phase stability. In this work, it is shown that these surfaces at small dimensions affect the relative equilibrium thermodynamics of the different phases. The CALPHAD approach for material surfaces (also termed "nano-CALPHAD") is employed to investigate these changes in three binary systems by calculating their phase diagrams at nano dimensions and comparing them with their bulk counterparts. The surface energy contribution, which is the dominant factor in causing these changes, is evaluated using the spherical particle approximation. It is first validated with the Au-Si system for which experimental data on phase stability of spherical nano-sized particles is available, and then extended to calculate phase diagrams of similarly sized particles of Ge-Si and Al-Cu. Additionally, the surface energies of the associated compounds are calculated using DFT, and integrated into the thermodynamic model of the respective binary systems. In this work we found changes in miscibilities, reaction compositions of about 5 at%, and solubility temperatures ranging from 100-200 K for particles of sizes 5 nm, indicating the importance of phase equilibrium analysis at nano dimensions.

Metal cluster-deposited graphene as an adsorptive material for m-xylene

m-Xylene weakly adsorbs on graphene and silver cluster doped graphene, but it has excellent interaction with platinum cluster doped graphene.