modeling studies. Entrenched in the development of these tools are critically derived and/or critically reviewed data sets used to develop and calibrate models to enable accurate prediction of a material’s physical properties.

With increasing ability to modify materials at the nano- and microscale, the need for next-generation materials databases is critical for industry adoption. In the design of a next-generation materials database, strong consideration should be given to fully understanding the queries that will be made against the database. This is also a dynamical issue in the sense that query response time is an important requirement. For instance, the Chemical and Biochemical Reference Dataset is much acclaimed due to its accuracy and completeness but it is implemented in an object-oriented database that makes it difficult to search across substructures.

In addition to measuring physical science data that are critical to the successful application of advanced materials and process models and simulations, NIST has undertaken the task of developing a database that would provide for the ready retrieval and updating of the data. This task has several challenges, which include the following: (1) the database can be quite large and heterogeneous, containing not only physical property data but associated metadata, such as microstructure photomicrographs, specification of testing parameters, and results of a simulation; (2) the same material is often given many different names, which presents an ontology challenge and the need to compile a thesaurus of synonyms; and (3) the database should implement a data model that is flexible and readily extensible in order to accommodate new data and legacy data that are stored in a variety of different database types.

In addition to its own modeling tools and generation of data, NIST could build on its connections to both industrial research laboratories and academic institutions as a means to supplement its measurements and the development of next-generation modeling tools. As data accessibility faces fewer technical barriers (e.g., utilizing cloud computing and big-data advances), databases could be designed to allow many input sources but would also require oversight to ensure consistency, completeness, and quality control. NIST has an important role to play in establishing the infrastructure for such databases and model dissemination, and in quality control.

Considering the current portfolio of materials characterization techniques now in use, it is apparent that physical property data—ranging from interatomic potential measurements, to detailed thermodynamic and kinetic models of diffusion, to precipitation modeling of complex alloy systems—are all under various stages of development and use. This critically important and unique collection of capabilities is the foundation for supporting future manufacturing initiatives, in which detailed constitutive models can link the computational materials, computational mechanics, and computational manufacturing models needed to derive materials by design and drive materials reinvention.

As is to be expected technologically, there is overlap between the areas of materials and manufacturing. With respect to the development of advanced manufacturing tools, the opportunity exists within NIST to collaborate with industrial manufacturing research, in which focused deliveries of the constitutive materials and mechanics models are key elements of next-generation computational manufacturing models. The key to the success of these NIST manufacturing programs is to move into new domains, in which manufacturing addresses the needs that include advances in stamping, casting, forging, and injection molding, and services the needs of the transportation industry, including automotive, aerospace, and rail, as well as infrastructure rebuilding and construction industries.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement