An official website of the United States government
Abstract Data‐driven science and technology have helped achieve meaningful technological advancements in areas such as materials/drug discovery and health care, but efforts to apply high‐end data science algorithms to the areas of glass and ceramics are still limited. Many glass and ceramic researchers are interested in enhancing their work by using more data and data analytics to develop better functional materials more efficiently. Simultaneously, the data science community is looking for a way to access materials data resources to test and validate their advanced computational learning algorithms. To address this issue, The American Ceramic Society (ACerS) convened a Glass and Ceramic Data Science Workshop in February 2018, sponsored by the National Institute for Standards and Technology (NIST) Advanced Manufacturing Technologies (AMTech) program. The workshop brought together a select group of leaders in the data science, informatics, and glass and ceramics communities, ACerS, and Nexight Group to identify the greatest opportunities and mechanisms for facilitating increased collaboration and coordination between these communities. This article summarizes workshop discussions about the current challenges that limit interactions and collaboration between the glass and ceramic and data science communities, opportunities for a coordinated approach that leverages existing knowledge in both communities, and a clear path toward the enhanced use of data science technologies for functional glass and ceramic research and development.
more »
« less
Machine learning‐based accelerated design of fluorphlogopite glass ceramic chemistries with targeted hardness
Abstract In this work, we develop and employ an accelerated design strategy using a machine learning algorithm to overcome the challenges for designing a new machinable glass ceramic. The trained machine learning model predicts the specific hardness value for numerous possibilities of processing conditions such as growth temperature and time. We report that the optimized growth parameters of 1200°C and 5 h achieve the highest machinability of 0.4 in the glass ceramic. Furthermore, we predicted the eight most promising candidates containing specific ratios of silicon, magnesium, aluminum, lithium, boron, potassium, barium, and oxygen. Combining machine learning with experimental data enables a systemic and rapid design of a ceramic material while capturing the underlying physics represented in the experimental data.
more »
« less
- Award ID(s):
- 2114595
- PAR ID:
- 10418060
- Publisher / Repository:
- Wiley-Blackwell
- Date Published:
- Journal Name:
- Journal of the American Ceramic Society
- Volume:
- 106
- Issue:
- 8
- ISSN:
- 0002-7820
- Page Range / eLocation ID:
- p. 4654-4663
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract The use of machine learning techniques to expedite the discovery and development of new materials is an essential step towards the acceleration of a new generation of domain-specific highly functional material systems. In this paper, we use the test case of bulk metallic glasses to highlight the key issues in the field of high throughput predictions and propose a new probabilistic analysis of rules for glass forming ability using rough set theory. This approach has been applied to a broad range of binary alloy compositions in order to predict new metallic glass compositions. Our data driven approach takes into account not only a broad variety of thermodynamic, structural and kinetic based criteria, but also incorporates qualitative and descriptive attributes associated with eutectic points in phase diagrams. For the latter, we demonstrate the use of automated machine learning methods that go far beyond text recognition approaches by also being able to interpret phase diagrams. When combined with structural descriptors, this approach provides the foundations to develop a hierarchical probabilistic predication tool that can rank the feasibility of glass formation.more » « less
-
Mechanistic data-driven prediction of as-built mechanical properties in metal additive manufacturingAbstract Metal additive manufacturing provides remarkable flexibility in geometry and component design, but localized heating/cooling heterogeneity leads to spatial variations of as-built mechanical properties, significantly complicating the materials design process. To this end, we develop a mechanistic data-driven framework integrating wavelet transforms and convolutional neural networks to predict location-dependent mechanical properties over fabricated parts based on process-induced temperature sequences, i.e., thermal histories. The framework enables multiresolution analysis and importance analysis to reveal dominant mechanistic features underlying the additive manufacturing process, such as critical temperature ranges and fundamental thermal frequencies. We systematically compare the developed approach with other machine learning methods. The results demonstrate that the developed approach achieves reasonably good predictive capability using a small amount of noisy experimental data. It provides a concrete foundation for a revolutionary methodology that predicts spatial and temporal evolution of mechanical properties leveraging domain-specific knowledge and cutting-edge machine and deep learning technologies.more » « less
-
Abstract This study explores the potential of single crystal barium titanate (BTO) platelets to fabricate nontoxic ceramics with enhanced material properties through texturization of grain structure. The proposed methodology relies on direct ink write additive manufacturing to enable grain‐oriented growth of BTO ceramics by utilizing a combination of spherical and platelet‐shaped particles. The use of platelet‐shaped particles in the ceramic ink guides particle alignment parallel to the build plate due to shear forces at the nozzle during the printing process. While platelet contents ranging from 0 to 40 wt.% showed a decrease in density as the content increased, experimental data revealed an incremental trend between platelet content, dielectric properties, and the degree of alignment of the particles on the F200 crystal plane, achieving a maximum texturized orientation of 65%. Such orientation resulted in 29.55% improved dielectric properties compared with randomly oriented BTO ceramic. The findings of this research validate the effectiveness of additive manufacturing technologies to tailor the microstructural characteristics of ceramics for specific functional applications.more » « less
-
Abstract This study employs a data‐driven machine learning approach to investigate specific ferroelectric properties of Al1−xScxN thin films, targeting their application in next‐generation nonvolatile memory (NVM) devices. This approach analyzes a vast design space, encompassing over a million data points, to predict a wide range of coercive field values that are crucial for optimizing Al1−xScxN‐based NVM devices. We evaluated seven machine learning models to predict the coercive field across a range of conditions, identifying the random forest algorithm as the most accurate, with a testR2value of 0.88. The model utilized five key features: film thickness, measurement frequency, operating temperature, scandium concentration, and growth temperature to predict the design space. Our analysis spans 13 distinct scandium concentrations and 13 growth temperatures, encompassing thicknesses from 9–1000 nm, frequencies from 1 to 100 kHz, and operating temperatures from 273 to 700 K. The predictions revealed dominant coercive field values between 3.0 and 4.5 MV/cm, offering valuable insights for the precise engineering of Al1−xScxN‐based NVM devices. This work underscores the potential of machine learning in guiding the development of advanced ferroelectric materials with tailored properties for enhanced device performance.more » « less
