Annual Report: NSF Award #EPS-1003897
I. Executive Summary
II. Detailed Report
A. RII participants and participating institutions
B. Program/Project Description
B. 1. RESEARCH ACCOMPLISHMENTS AND PLANS:
B.1.1. Science Driver 1: Electronic and Magnetic Materials
B.1.2. Science Driver 2: Energy Materials
B.1.3. Science Driver 3: Biomolecular Materials
B.2. DIVERSITY AND BROADENING PARTICIPATION, INCLUDING INSTITUTIONAL COLLABORATIONS
B.3. WORKFORCE DEVELOPMENT:
B.4. CYBERINFRASTRUCTURE:
B.5. EXTERNAL ENGAGEMENT:
B.6. EVALUATION AND ASSESSMENT:
B.7. SUSTAINABILITY AND PROJECT OUTPUTS:
C. Management Structure:
D. Jurisdictional and Other Support:
E. Planning Updates:
F. Unobligated Funds:
G. Progress with Respect to the RII Strategic Plan:
H. Jurisdiction Specific Terms and Conditions:
I. Reverse Site Visit (RSV) Recommendations:
J. Experimental Facilities:
K. Publications and Patents:
L. Honors and Awards:
References
I. Executive Summary (to be revised by PET)
Introduction. This NSF award has led to the establishment of the Louisiana Alliance for Simulation-Guided Materials Applications (LA-SiGMA), which brings together Louisiana academic institutionsto focus on simulation-assisted materials by design in research and in the education of a new generation of Louisiana material scientists. Three “Science Driver” areas have been identified for specific focus: (1) correlated electronic materials, which show promise as new materials for the design of molecular computers, microelectronics, and high-density recording media; (2) energy materials, which show promise as catalysts, advanced materials for the storage and release of hydrogen, and electrochemical cells and capacitors that store and deliver electrical energy; and (3) biomolecular materials, which can provide encapsulation, delivery, and release of therapeutics to specified targets. LA-SiGMA takes advantage of the Louisiana Optical Network Initiative (LONI), the most advanced high performance computing, network, and communication infrastructure among EPSCoR states. Since the principal barrier to simulation-guided design of materials—multiple length and time scales—challenges conventional scientific disciplines, LA-SiGMA combines researchers with specialized expertise at each scale into teams. Teams include applied mathematicians and computer scientists to ensure efficient utilization of forthcoming high performance computers and include experimentalists to test, validate and guide the computational simulations. These collaborations will position Louisiana to compete effectively for a national center of excellence in multiscale materials modeling and simulation.
Vision: Transformative advances in materials science research and education through a sustained multidisciplinary and multi-institutional alliance of researchers.
Mission: to establish a sustained national center of excellence in computational material science by the end of the project period.
Goals:
Science Driver 1: Electronic and Magnetic Materials: Transform the field by extending many-body formalisms and first principles methods to much larger length scales than currently possible.
Science Driver 2: Materials for Energy Storage and Conversion: Develop and apply multi-scale computational tools to study materials for energy generation, storage, and conversion.
Science Driver 3: Biomolecular Materials: Develop, apply, and validate experimentally multi-scale computational tools for the design of novel vehicles for drug delivery and other applications.
Computational Teams: Develop multi-scale formalisms, algorithms, and codes for materials simulations and modeling; leverage existing tools and make optimum use of the next generation computing environments.
Research Efforts. LA-SiGMA consists of three Science Drivers (SD) and a cyberinfrastructure and cybertools team (CTCI) which incorporates three computational teams with membership rich in Science Driver participants. The structure of the CTCI team enables the development of common computational research tools.
Science Driver 1: Electronic and Magnetic Materials. The goals of the Science Driver 1 (SD1) team are to develop and validate methods that enable the study of complex phenomena in correlated electronic and magnetic materials ranging from transition metal oxides to organic magnets. Complex emergent phenomena include properties such as superconductivity, magnetic, charge or orbital ordering, or other phases that one cannot predict even with an exact and complete understanding of the constituent atoms. These phases often compete, making these materials very sensitive to applied fields and perturbations that can lead to new functionalities. As a consequence of their exotic and diverse physical properties, transition metal oxides are considered to be the frontier of research on “emergent research device materials.” To study these systems, the SD1 team is developing mulitscale methods able to treat physics on the different length scales which characterize these orders, non-local density functional theory methods able to treat the strong correlation effects, and methods which combine these two techniques. The SD1 team includes researchers at LATech, LSU, Southern, Tulane, Xavier and UNO, with a high degree of overlap with the CTCI team. Over 30 students and postdocs have been recruited into SD1.
Science Driver 2: Materials for Energy Storage and Generation. The goal of Science Driver 2 (SD2) is to build a molecular level understanding of materials of importance for energy storage and generation. This understanding is important for improving these materials to meet global energy challenges. The work targets electrode materials for supercapacitors and hydrogen storage materials, and is developing molecular models to investigate catalytic process for the formation of biofuels and toxic biproducts from combustion. The team’s computational goals are to bring insight into the described systems by utilizing accurate ab initio methods, while developing the computational methodology to link these insights—which are often on the scale of a few to dozens of atoms—to the macroscopic process being investigated. This requires the development of new force fields that allow for the simulation of 10,000s of atoms that are parameterized by accurate ab initio results. Furthermore, there is a need to link with longer timescales that are outside the purview of atomistic simulations. Hence, the SD2 team is developing a mathematical model that can link the molecular level properties with long scale dynamical variables.
Science Driver 3: Biomolecular Materials. The goals of Science Driver 3 (SD3) team are to develop, apply, and validate experimentally multi-scale computational tools that will enable the design of novel drug delivery vehicles. The barriers to achieving these goals are the complexity associated with modeling systems with atomistic details over length scales on the order of 10-9 to 10-7 m and the lack of sufficiently efficient force fields to enable simulations to reach time scales of 10-6 to 10-3 s is required for meaningful predictions. SD3 is composed of multi-disciplinary teams of scientists at four institutions (Tulane, LSU, UNO, and LA Tech) that combine theoretical/computational scientists with experimentalists to investigate aspects necessary for building predictive models of uni-molecular and multi-molecular vehicles for targeted drug delivery.
Cyberinfrastructure, Cybertools and Computational Teams (CTCI). CTCI is focused on the development of the formalisms, algorithms and codes used in this project needed to efficiently utilize the next generation of supercomputers. A major research goal of CTCI is to collaborate with the computational teams: (1) Next Generation Monte Carlo Codes, (2) Massively Parallel DFT and Force Field Methods, and (3) Large-scale Molecular Dynamics. In addition, CTCI is also working on execution management, data management, visualization and heterogeneous (GPU) computing. CTCI includes members of all of computational and SD teams. CTCI is the glue that holds the three SDs together.
Diversity. The Diversity Advisory Council is coordinating existing and planned efforts to meet the project’s diversity goals. One major accomplishment, for instance, is the establishment of the Supervised Undergraduate Research Experience program, which is restricted to women and underrepresented minorities. Thirty (30) Louisiana students were competitively selected out of 127 applicants to work with faculty mentors at universities around the State, and future competitions are planned.
Workforce Development. The project contributes to workforce development in the State through the creation of a multi-tiered set of educational programs for graduate and postgraduate students who will enter the workforce as well-trained computational materials scientists. Four graduate level courses have been offered using HD synchronous video. Additional programs encourage high school, community college, and undergraduate students to explore computational materials science as a career, including Research Experiences for Teachers summer programs in New Orleans, Baton Rouge, and Ruston, as well as Research Experiences for Undergraduates (REU) programs at six LA-SiGMA universities.
External Engagement. LA-SiGMA maintains a web portal, a Facebook page, and an SVN (a collaborative software and revision system that maintains current and historical versions of files such as source code, web pages, and documentation). Synchronous collaboration tools such as EVO enable inter-institutional collaboration. The LA EPSCoR monthly newsletter continues to highlight the role played by LA EPSCoR in promoting the development of the State’s S&T resources through partnerships involving its universities, industry, and government. The newsletter is distributed to approximately 1,000 individuals, including faculty members, State legislators, and other stakeholders. The Speaking of Science (SoS) speaker’s bureau introduces thousands of K-12 students to the exciting science being undertaken by Louisiana’s best researchers. During this reporting period, 60 presentations were given to over 2,200 students, an audience comprised of 49% females and 38% underrepresented minorities. Two industry/academia collaborative workshops were held, with the goal of fostering interactions between Louisiana’s leading researchers and representatives of Louisiana’s industrial Research & Development community. One focused on the theme of energy and materials science (100 attendees) and the other (60 attendees) centered on Digital Media and Software Development, a rapidly growing industry in Louisiana.
Evaluation and Assessment. Multiple layers of evaluation and assessment is available to LA-SiGMA: An internal Evaluation and Assessment team, an external evaluator, and the External Review Board (ERB). The Office of Educational Innovation and Evaluation (OEIE) of Kansas State University continues to act as the external evaluator for Louisiana's EPSCoR programs. The logic models and strategic plan developed during Year 1 and accepted by the NSF were used to design and implement an on-line data collection portal called the "Online Advancing Science Information System (OASIS)." Launched towards the end of Year 1, OASIS was used extensively for the data collection needed for the Year 2 report. The data collection templates in OASIS were designed specifically to provide the data requested in the NSF templates and also to help the evaluators assess LA-SiGMA's progress in meeting the strategic plan milestones and outcomes.
Sustainability. A host of initiatives for improving the State’s research infrastructure and enhancing sustainability are administered by the LA EPSCoR office. These programs include: Pilot Funding for New Research ; Links with Industry, Research Centers and National Labs ; Opportunities for Partnerships with Industry ; Preliminary Planning Grants for Major Initiatives ; Travel Grants for Emerging Faculty ; Grantwriting Workshops ; the LA Genius Faculty Expertise Database ; and the SBIR/STTR Phase Zero program. All of these programs have a significant impact on propelling Louisiana’s S&T enterprise to become more competitive and sustainable. LA-SiGMA is actively focused on development of structures that will sustain it well beyond the end of the funding. LA-SiGMA members have led the effort to obtain a DOE Predictive Theory and Modelling Center which would bring the development of NWChem to the state and establish a significant collaboration with Pacific Northwest National Lab. We hosted an NWChem training workshop attended by ??? students. We have taught and are developing graduate level courses to train the current and future generation of computational materials scientists. LA-SiGMA is leveraging its funding and equipment purchases to obtain additional funding and equipment based on our successes. LA-SiGMA is developing new national and international collaborations.
Management Structure. Dr. Michael Khonsari is the Project Director (PD), and Mr. Jim Gershey is Project Administrator (PA). The PD ensures that the various stakeholders operate as a cohesive research enterprise progressing towards realization of project goals and objectives. A 21-member EPSCoR Committee meets at least twice a year. Assisting Dr. Khonsari and the EPSCoR Committee is a professional staff of five full-time individuals whose responsibilities include program management, fiscal and contract management, database administration, coordination of statewide outreach, and communications. The Project Execution Team (PET) oversees the day-to-day activities of the LA-SiGMA project and provides direction and guidance to project participants in each of the science driver and computational teams. The management structure features several interconnected teams designed to effectively implement and assess the project goals, promote project-wide participation in leadership, and ensure effective communications. These teams include the following: Science & Cyberinfrastructure Team; External Engagement & Workforce Development Team; Evaluation & Assessment Team; Industrial Liaison Team (ILT); and Diversity Advisory Council (DAC). An External Review Board (ERB), consisting of a diverse group of eminent scholars, conducts comprehensive programmatic reviews and forwards objective guidance, feedback, and recommendations to the PD and PET, to ensure that program goals and objectives are being met. Biannual “all-hands” face-to-face meetings are used for coordination and identification of new collaborations. The DAC and ILT participate in the biannual all-hands meetings. The ERB attends at least one of these biannual meetings to evaluate all aspects of the project.
Key Accomplishments. Intellectual Merit: New technologies often depend on designing new materials for specific tasks. Examples include materials for computer memories, batteries, and controlled drug delivery. When developing new materials, scientists and engineers search through a vast range of possibilities including complicated mixtures of ingredients, or architectural organizations of materials that present themselves in structures of different sizes and shapes. For example, a material for controlled drug delivery may have to be porous enough to absorb a drug but have a coating that protects the contents until the release environment is encountered. Design of useful, cost effective, and environmentally friendly new materials is a grand challenge for materials scientists. This challenge requires understanding, modeling, and exploiting the complexity of materials on numerous levels, from the small scale of minute constituents to the grand scale of technological applications. The interplay of behavior on multiple time and length scales (i.e., multiscale behavior) can lead to unexpected complex emergent phenomena. Such problems are generally unsolved in materials science.
Modern methods of experimental science and engineering heighten the design challenge by offering the ability to produce designed materials and to test detailed design features, though testing all possibilities is usually not economically feasible. Paralleling this expansion of experimental methods has been an explosion in the power of modern computers, and the evolution of the sophisticated computational science algorithms for simulation of materials, particularly specialized algorithms that take advantage of the hardware advances. The combined progress suggests the goal of simulation-assisted materials by design. This goal has been widely appreciated, and steady progress continues, but transformative progress will require a confluence of experimental and computational facilities together with directed intellectual collaboration. This RII research project builds upon existing state and federal investments in experimental and computational facilities in Louisiana.
Broader Impact : LA-SiGMA is effecting a transformative and sustainable change in computational materials research, education, and applications throughout the State of Louisiana and making the State competitive for a federally funded research center. LA-SiGMA is pushing the scientific frontiers in computational materials science, and enabling Louisiana researchers to use the next generation of heterogeneous, multicore and hyperparallel cyberinfrastructure effectively. LA-SiGMA is building statewide interdisciplinary research collaborations involving computational scientists, computer scientists and engineers, applied mathematicians, theorists and experimentalists. Most significantly, the project will build an inter-institutional computational materials science graduate program that will be unique in its statewide reach and impact, and may be the only one of its kind in the nation.
SD1: Research highlights include a partnership with CTCI to develop massively parallel GPU enhanced codes. These codes employ the state's GPU supercomputers to enable new discovery, and are also ready for use on BlueWaters. John Perdew and his collaborators are developing greatly improved semilocal and nonlocal density functionals that will be employed in electronic structure calculations for molecular magnets and other complex materials of interest to the SD1 science driver team. Experiments are used to both validate our methods and to suggest new materials for study. They are being performed on a wide spectrum of materials, ranging from iron-based high temperature superconductors to organic magnets. SD1 researchers have led the way in teaching four semester-long distance learning courses open to all LA-SiGMA participants, as well as numerous researchers in Europe and North America, teaching methods directly relevant to the research being done in SD1 and CTCI. Our efforts are coordinated through bi-weekly meetings of SD1 and of the GPU team in which SD1 is heavily involved.
SD2: One of the stated milestones is to develop algorithms that combine chemistry and physics of energy storage. During the second year, the SD2 team has carried out calculations to understand how charge is carried and distributed in the electrochemical double-layer capacitors, which are potential components for supercapacitors. Simulations have also been performed to understand how the addition of impurities to a sodium magnesium hydride can increase its ability to store hydrogen, and investigated how adding impurities to C60 influence its thermoelectric properties. The team has carried out DFT calculations to extract appropriate force field parameters to study catalysis. Furthermore, the team has moved into an exciting new area of Lithium-ion battery research, which combines experimental real time imaging combined with DFT calculations. The team’s efforts are coordinated via monthly synchronous meetings using the EVO collaboration tool and strong collaborations with the force field methods and Monte Carlo computational teams with CTCI.