Machine-Learning-Based phase diagram construction for high-throughput batch experiments

Deffrennes, Guillaume; Han, Kwangsik; Abe, Taichi; Morito, Haruhiko; Nakamura, Yasuyuki; Naito, Masanobu; Katsube, Ryoji; Nosé, Yukihiko; Terayama, Kei

doi:10.1080/27660400.2022.2076548

Cited by 4 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…48 Another consideration when choosing an acquisition function is whether experiments are performed individually or in batches. [49][50][51][52] When experiments are performed in batches, it is necessary to take action to make sure that each point will target a different region in x. One approach, for instance, is to select the rst experiment in the batch, then recondition the GPR assuming that the point was sampled and either returned the GPR mean (Kriging believer) or a constant value (constant liar), and use this updated GPR to predict the next point iteratively until all batch points have been selected.…”

Section: B the Metrics That Dene Performancementioning

confidence: 99%

Driving school for self-driving labs

Snapp,

Brown

2023

Digital Discovery

View full text Add to dashboard Cite

show abstract

Section: B the Metrics That Dene Performancementioning

confidence: 99%

Driving school for self-driving labs

Snapp,

Brown

2023

Digital Discovery

View full text Add to dashboard Cite

show abstract

“…As opposed to the DoE, machine-learning techniques can handle both mixed and categorical variables and synthesis outcomes (Scheme ). This renders them better suited for problems involving the synthesis of new materials across a wide range of crystal phases; for example, Tamura et al recently considered strategies to effectively construct phase diagrams using high-throughput batch experiments . For synthetic problems where high-throughput synthesis is not plausible, data-driven classifiers can train on relatively small data sets to render synthetic phase maps that allow for the rational targeting of materials within a high-dimensional parameter space .…”

Section: Doe Versus Machine Learningmentioning

confidence: 99%

“…This renders them better suited for problems involving the synthesis of new materials across a wide range of crystal phases; for example, Tamura et al recently considered strategies to effectively construct phase diagrams using high-throughput batch experiments. 47 For synthetic problems where high-throughput synthesis is not plausible, data-driven classifiers can train on relatively small data sets to render synthetic phase maps that allow for the rational targeting of materials within a high-dimensional parameter space. 48 The efficacy of the experimental training data sets can be maximized by utilizing design matrixes typically seen in the DoE to rationally sample an n -dimensional design space and increase its robustness.…”

Section: Doe Versus Machine Learningmentioning

confidence: 99%

Using Data-Driven Learning to Predict and Control the Outcomes of Inorganic Materials Synthesis

Williamson,

Brutchey

2023

Inorg. Chem.

View full text Add to dashboard Cite

The design of inorganic materials for various applications critically depends on our ability to manipulate their synthesis in a rational, robust, and controllable fashion. Different from the conventional trial-and-error approach, data-driven techniques such as the design of experiments (DoE) and machine learning are an effective and more efficient way to predictably control materials synthesis. Here, we present a Viewpoint on recent progress in leveraging such techniques for predicting and controlling the outcomes of inorganic materials synthesis. We first compare how the design choice (statistical DoE vs machine learning) affects the type of control it can offer over the resulting product attributes, information elucidated, and experimental cost. These attributes are supported by discussing select case studies from the recent literature that highlight the power of these techniques for materials synthesis. The influence of experimental bias is next discussed, followed finally by our perspectives on the major challenges in the widespread implementation of predictable and controllable materials synthesis using data-driven techniques.

show abstract

“…The Materials Genome Initiative accelerated materials discovery through efforts such as the Materials Project, which combines supercomputing and density functional theory (DFT) to theoretically predict new materials and their properties before they are made. , While this “materials by design” approach successfully identified vast numbers of materials with a wide range of targeted properties, a significant bottleneck now exists at the next step of the process: the Edisonian nature of materials synthesis. Unlike the vast majority of related reports from previous studies that develop approaches to discover new materials with specific properties, − there is no robust predictive framework that can help map the reaction coordinate from precursors to the final crystalline solid when attempting to synthesize materials. Moreover, the compositions and structure types of crystalline inorganic solids are so disparate that it is exceptionally challenging to apply the lessons learned from one materials system to another.…”

Section: Introductionmentioning

confidence: 99%

Predictive Synthesis of Copper Selenides Using a Multidimensional Phase Map Constructed with a Data-Driven Classifier

et al. 2023

View full text Add to dashboard Cite

Copper selenides are an important family of materials with applications in catalysis, plasmonics, photovoltaics, and thermoelectrics. Despite being a binary material system, the Cu–Se phase diagram is complex and contains multiple crystal structures in addition to several metastable structures that are not found on the thermodynamic phase diagram. Consequently, the ability to synthetically navigate this complex phase space poses a significant challenge. We demonstrate that data-driven learning can successfully map this phase space in a minimal number of experiments. We combine soft chemistry (chimie douce) synthetic methods with multivariate analyses via classification techniques to enable predictive phase determination. A surrogate model was constructed with experimental data derived from a design matrix of four experimental variables: C–Se bond strength of the selenium precursor, time, temperature, and solvent composition. The reactions in the surrogate model resulted in 11 distinct phase combinations of copper selenide. These data were used to train a classification model that predicts the phase with 95.7% accuracy. The resulting decision tree enabled conclusions to be drawn about how the experimental variables affect the phase and provided prescriptive synthetic conditions for specific phase isolation. This guided the accelerated phase targeting in a minimum number of experiments of klockmannite CuSe, which could not be isolated in any of the reactions used to construct the surrogate model. The reaction conditions that the model predicted to synthesize klockmannite CuSe were experimentally validated, highlighting the utility of this approach.

show abstract

Machine-Learning-Based phase diagram construction for high-throughput batch experiments

Cited by 4 publications

References 33 publications

Driving school for self-driving labs

Driving school for self-driving labs

Using Data-Driven Learning to Predict and Control the Outcomes of Inorganic Materials Synthesis

Predictive Synthesis of Copper Selenides Using a Multidimensional Phase Map Constructed with a Data-Driven Classifier

Contact Info

Product

Resources

About