Role of cyberinfrastructure in educating the next generation of computational materials scientists

  • Susan B SinnottEmail author
  • Simon R Phillpot
Open Access
Part of the following topical collections:
  1. Use of Digital Data in Materials Science and Engineering


An overview of cyberinfrastructures developed to advance the field of materials modeling is presented. The role of cyberinfrastructures in educating the next generation of the workforce is also discussed, with an emphasis on the Cyberinfrastructure for Atomistic Simulation (CAMS). The paper concludes with a summary regarding the future outlook of cyberinfrastructures, especially with regard to education.


Cyberinfrastructure Material modeling Education 


The last decade has witnessed tremendous growth in the development and utilization of cyberinfrastructures in a variety of science and engineering disciplines. This includes the materials modeling community, where cyberinfrastructures have been used to enable the sharing of information to advance a specific objective within a team of researchers, or within a specific community to distribute information or establish best practices. For example, the Atomic-scale Friction Research and Education Synergy Hub (AFRESH) [1] was developed to share links to computational tools, data provided by users and mined from the literature, and best practices for modeling friction with atomic-scale simulations.

Cyberinfrastructures have additionally been used to host databases generated from a particular experimental apparatus, from a computational program, and/or from published literature. Having extensive, specialized data gathered in a single location and available to other researchers is extraordinarily beneficial for subsequent material design and discovery. This has therefore been a common way in which cyberinfrastructure has been used, at least in part. For instance, the Materials Project at MIT [2] provides access to databases of material properties generated using first-principles calculations. The focus is primarily on materials for energy storage applications, although the information may be useful for a variety of applications. Another example is the AFLOWLIB [3] cyberinfrastructure, hosted at Duke University, that provides access to an extensive database of metal alloy properties and phase diagrams. AFLOWLIB also hosts databases of material properties that include structural, electronic, and thermoelectric characteristics, to name just a few.

An important use of cyberinfrastructures for computational materials science has been to give broad access to computational tools of various types. For example, in the case of AFRESH access to tools for displaying the output of atomic-scale simulations and examples for their optimal use were provided. Additionally, both the Materials Project and AFLOWLIB allow users to carry electronic structure calculations through the cyberinfrastructure itself at computers located at the hosting institution. Cyberinfrastructures may further serve to organize and disseminate information that is critical for the optimal use of a computational method. This is exemplified by the openKIM project [4] and the NIST Interatomic Potentials Repository Project [5] that catalog interatomic potentials for classical simulations in a unique way so that they may be clearly identified when they are used. These cyberinfrastructures further host their organizational systems in a user friendly way and propagate information to the community as a whole.

An additional important application of cyberinfrastructure in computational materials science and engineering is to enhance the education of the next generation of computational materials scientists. For example, the openKIM project has held several workshops, primarily students and postdocs, about the openKIM cataloging system, including information on how to migrate potentials to the cyberinfrastructure, and to how to fully used the of the openKIM system for potential development.

This educational role is critical and will deepen the understanding of newcomers to the field, assist them with learning to use important technical software for computing and data analysis, and provide a forum for them to share information. Educating the next generation in this manner is also a critical step to ensure the long-term viability of the cyberinfrastructure itself. Indeed, it will ultimately be this next generation who will use, maintain, and expand these platforms.


The Cyberinfrastructure for Atomistic Materials Science (CAMS) [6] is a pilot-project stage platform focused on modeling materials at the atomic scale, with an emphasis on microstructural features. It fills a number of the roles discussed above. For example, it houses an actively curated “virtual library” of microstructure samples, where each sample contains the atomic-scale coordinates of a specified material microstructure. The microstructures that are featured include multiple grain boundaries, surfaces, dislocations, nanocrystalline samples, and related structures. CAMS also links to similar databases to assist the user in finding the structure of interest, reduce duplication across such libraries, and ultimately to encourage the sharing of information across platforms. Users are also encouraged to contribute samples to the CMAS library.

Importantly, CAMS is actively engaged in educational activities (see Figure 1). Part of the motivation for this is to increase use of the virtual library and to grow the number of samples that are contained within it. However, the main motivation is tied to the central mission of CAMS, which is to assist the materials community by serving as a hub for disseminating information on atomic-scale modeling and related methods, especially as they apply to microstructural features within materials. A key target constituency is the next generation of materials modelers, including students, postdocs, junior faculty, and other newcomers to the field.
Figure 1

Information page for CAMS summer school.

To reach this audience, in spring 2013 CAMS sponsored and hosted a summer school on “Simulation of Complex Microstructure in Materials” at the University of Florida. More than 50 participants from 20 different U.S. institutions and one German university participated. The instructors for the week-long school were computational and experimental researchers with considerable expertise from multiple U.S. academic institutions, one German university, and U.S. National Laboratories. After each presentation, participants were charged in working in groups to develop questions for the speakers. In so doing, the group members answered many of each other’s questions and reinforced the learning that had taken place in the presentation. The lecturer then returned for an extended question and answer session; the questions that the groups then brought forward were sophisticated and well thought out, and thus initiated extensive and detailed further discussions. Many participants, both students and lecturers, commented on the effectiveness of this approach. In addition, the participants were given a training session on how to use and upload structures to the CAMS virtual library; they were further encouraged to contribute structures to it during and after the school. The participants presented research results from their home institution in a poster session and worked together in teams to develop research proposals based on the topics presented at the school. Awards were given to the best posters and best proposals by a panel of judges populated by a diverse mix of instructors. Lastly, the school set aside time for professional development activities, including a discussion led by a representative of the University of Florida’s College of Engineering Entrepreneurship Institute, and discussions of their professional lives by a faculty member at a research-intensive institution and a staff member at a US Department of Energy National Laboratory.

The CAMS-run summer school thus contained many opportunities for technical, professional, and interpersonal development on the part of the participants. While travel grants to the participants ensured broad participation, the number of attendees was still limited by the inevitable logistical limitations of any such activity. Therefore, copies of most of the presentations and videos of most of the lectures are provided free of charge on CAMS (see Figure 2) to anyone who goes through a simple registration process. In this way, the impact of the educational effort is disseminated broadly thus further increasing accessibility.
Figure 2

Posted presentations for the 2013 CAMS summer school.

In this respect CAMS joins other, more established cyberinfrastructures that are focused on educational activities. For instance, the CAVS CyberDesign effort at Mississippi State University [7] is a comprehensive wiki site devoted to the topic of integrated computational material engineering. Specifically, this cyberinfrastructure is focused on providing access and guidance to computational materials science methods to enable their use in conjunction with design and manufacturing. Several different models are provided related to materials design and educational courses at a variety of educational institutions, highlighting the power of a cyberinfrastructure to enable the sharing of expertise and educational tools among educators. Similarly, the nanoHUB [8] based at Purdue University contains a large number of computational tools, lectures, and other resources in the area of nanoscience and nonotechnology, with a particular strength in the area of electronic devices.


The future of cyberinfrastructures relies on their ability to successfully meet the needs of their constituencies and to provide resources that are deemed to be valuable. Additionally, resources to sustain their maintenance and continued growth have to be available. The ability of cyberinfrastructures to utilize each other’s resources is particularly key, as this will allow each one to magnify the effect of the individual effort. It is also important that they work with other, perhaps less formal, efforts in community building. Efforts such as the Materials Innovation @TMS website ( greatly assist cyberinfrastructures to connect with one another and achieve this goal.

An example of such cooperation is the spring 2014 summer school titled “Transformational Technologies in Molecular Simulations” that will be held at the University of Wisconsin-Madison (UW). This summer school is co-sponsored by CAMS and the MaterialsHUB [9], a cyberinfrastructure focused on the development of new theories and computational tools for the rapid calculation of material properties at the atomic scale. The UW Materials Research Science and Engineering Center (MRSEC) Interdisciplinary Computational Group is also co-sponsoring the summer school. This collaborative interaction promises a wider breadth of topics within the common focus area of interest to these cyberinfrastructures and interdisciplinary group. Ultimately, the collaboration should thus enrich the educational experience of the next generation computational materials scientists that participate in the school or access the subsequent electronically disseminated content.

The common initiatives among the cyberinfrastructures described here to catalog, curate, and distribute content from developers, users, and researchers within the field in such a way that each contributor to the cyberinfrastructure receives full credit is an important advance that, if adopted by the next generation of material modelers, will lead to a number of benefits. These include a new avenue to disseminate and to develop computational tools, to clarify the literature for everyone but especially to newcomers, and to ultimately lower the barrier for entry into the field of computational materials science by the non-expert.

The importance of materials modeling cyberinfrastructures is only expected to increase as the use of mobile electronic devices become more widespread, computational methodologies become a common approach in the material structure–property relationship toolbox, and the next generation of the workforce uses these platforms as an expected part of the modern research endeavor.



The authors gratefully acknowledge the support of the National Science Foundation (DMR-1246173).

Supplementary material

40192_2013_11_MOESM1_ESM.jpg (174 kb)
Authors’ original file for figure 1
40192_2013_11_MOESM2_ESM.jpeg (172 kb)
Authors’ original file for figure 2


  1. 1.
    Sinnott SB, Fortes JAB, Bucholz EW, Matsunaga AM: Atomic-scale Friction Research and Education Synergy. In NSF Engineering Research and Innovation Conference. Honolulu, Hawaii; 2009.Google Scholar
  2. 2.
    Jain A, Ong P, Hautier G, Chen W, Richards W, Dacek S, Cholia S, Gunter D, Skinner D, Ceder G: The materials project: a materials genome approach to accelerating materials innovation. App Phys Lett Materials 2013, 1: 011002.Google Scholar
  3. 3.
    Curtarolo S, Setyawan W, Wang S, Xue J, Yang K, Taylor R, Nelson L, Hart G, Sanvito S, Buongiorno-Nardelli M, Mingo N, Levy O: AFLOWLIB.ORG: A distributed materials properties repository from high-throughput Ab Initio calculations. Comput Mater Sci 2012, 58: 227–235.CrossRefGoogle Scholar
  4. 4.
    Tadmor EB, Elliott RS, Sethna JP, Miller RE, Becker CA: Knowledgebase of Interatomic Models (KIM). 2011. Available via https://openkim.orgGoogle Scholar
  5. 5.
    Becker CA, Tavazza F, Trautt ZT, de Macedo RA B: Considerations for choosing and using force fields and interatomic potentials in materials science and engineering. Curr Opinion Solid State Mater Sci 2013, 17: 277–283. Available via Scholar
  6. 6.
    Cyberinfrastructure for Atomistic Materials Science (CAMS). 2013. Available via
  7. 7.
    CAVS Engineering Virtual Organization for CyberDesign. 2013. Available via
  8. 8.
    Madhavan K, Zentner L, Farnsworth V, Shivarajapura S, Zentner M, Denny N, Klimeck G: Cloud-based services for nanoscale modeling, simulation, and education. Nanotechnology Reviews 2013, 2: 107–117. Available via https://nanoHUB.orgCrossRefGoogle Scholar
  9. 9.
    MaterialsHUB. 2013. Available via

Copyright information

© Sinnott and Phillpot; licensee Springer. 2014

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 2.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Authors and Affiliations

  1. 1.Department of Materials Science and EngineeringUniversity of FloridaGainesvilleUSA

Personalised recommendations