Sangtae Kim (Vice Chair)
This document is sponsored by the National Science Foundation (NSF) under grant No. ENG-0423742 to the World Technology Evaluation Center, Inc. The Government has certain rights in this material. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the United States Government, the authors’ parent institutions, or WTEC, Inc.
As a draft document, any conclusions are subject to revision.
Copyright 2007 by WTEC, Inc. Copyrights are reserved by individual authors or their assignees except as noted herein. The U.S. Government retains a nonexclusive and nontransferable license to exercise all exclusive rights provided by copyright. Some WTEC final reports are distributed by the National Technical Information Service (NTIS) of the U.S. Department of Commerce. A list of available WTEC reports and information on obtaining them is on the inside back cover of this report.
World Technology Evaluation Center, Inc. (WTEC)
R. D. Shelton, President
Michael DeHaemer, Executive Vice President
Geoffrey M. Holdridge, Vice President for Government Services
Halyna Paikoush, Event Planner and Administrative Support
Table of Contents
Appendix B. Site Reports—Asia
Site: Central Research Institute of Electric Power Industry (CRIEPI) 1
Site: Dalian University of Technology 7
Site: Institute for Molecular Science (IMS) 17
Site: Institute of Computational Mathematics and Scientific/Engineering Computing 20
Site: Institute of Process Engineering, Chinese Academy of Sciences 22
Site: Japan Agency for Marine-Earth Science and Technology
Earth Simulator Center (ESC) 25
Site: Kyoto University 27
Site: Mitsubishi Chemical Group Science and Technology Research Center (MCRC) 31
Site: Nissan Research Center, Fuel Cell Laboratory 34
Site: Peking University Center for Computational Science and Engineering 37
Site: Research Institute for Computational Sciences (RICS)
National Institute of Advanced Industrial Science and Technology (AIST) 39
Site: RIKEN – The Institute of Physical and Chemical Research
Advanced Center for Computing and Communication (ACCC) 42
Site: Shanghai Supercomputer Center 45
Site: Shanghai University 47
Site: The Systems Biology Institute (SBI) 50
Site: Toyota Central R&D Labs, Inc. 53
Site: Tsinghua University Department of Engineering Mechanics 56
Site: University of Tokyo 58
Site: Autonomous University of Barcelona and
Materials Science Institute of Barcelona (ICMAB-CSIC)
Research Center for Nanoscience and Nanotechnology 61
Site: BASF – The Chemical Company 65
Site: Center for Atomic-Scale Materials Design (CAMD)
Technical University of Denmark Department of Physics 70
Site: CERN (European Organization for Nuclear Research) 73
Site: CIMNE (International Center for Numerical Methods in Engineering) 76
Site: Ecole Polytechnique Fédérale de Lausanne (EPFL)
Institute of Analysis and Scientific Computing (IACS) 78
Site: Ecole Polytechnique Fédérale de Lausanne (EPFL), Blue Brain Project 82
Site: Eni SpA 83
Site: ETH (Swiss Federal Institute of Technology) Zürich 85
Site: Fraunhofer Institute for the Mechanics of Materials (IWM) 91
Site: IBM Zurich Laboratory, Deep Computing 93
Site: Institute Français du Petrol (French Petroleum Institute) 97
Site: Institute of Fluid Mechanics of Toulouse (IMFT) 102
Site: IRIT (Institut de Recherche en Informatique de Toulouse), and
ENSEEIHT (Ecole Nationale Supérieure d´Electrotechnique, d´Electronique, d´Informatique, d´Hydraulique et des Télécommunications) 104
Site: Paris Simulation Network 106
Site: Science & Technology Facilities Council (STFC) Daresbury Laboratory 110
Site: Technical University of Denmark (DTU) Wind Engineering
Department of Mechanical Engineering (MEK) 117
Site: Technical University of Denmark Center for Biological Sequence Analysis 120
Site: Technical University of Munich (Institut für Informatik), and
Leibniz Supercomputing Centre 125
Site: Unilever Centre for Molecular Informatics
University of Cambridge 135
Site: Unilever R&D Port Sunlight 137
Site: University College London 140
Site: University of Cambridge Centre for Computational Chemistry 145
Site: University of Cambridge Dept. of Applied Mathematics and Theoretical Physics (DAMTP) 147
Site: University of Cambridge Theory of Condensed Matter Group 150
Site: University of Karlsruhe and
Forschungszentrum Karlsruhe (Karlsruhe Research Center)
Karlsruhe Institute of Technology (KIT) and other affiliated institutes 153
Site: University of Oxford Condensed Matter Theory Group
Rudolf Peierls Centre for Theoretical Physics 156
Site: University of Oxford
Department of Engineering Science 158
Site: University of Oxford Theoretical Chemistry Group 162
Site: University of Oxford, Structural Bioinformatics
and Computational Biochemistry Group 164
Institute of Thermodynamics and Thermal Process Engineering 166
Site: University of Zurich Physical Chemistry Institute
Computational Chemistry Group of Prof. Dr. Jürg Hutter 173
Site: Vrije University Amsterdam
Dept. of Molecular Cell Physiology and
BioCentrum Amsterdam, Faculty of Biology 177
Site: Vrije University Theoretical Chemistry Section 181
Site: Zuse Institute Berlin (ZIB) 185
Appendix B. Site Reports—Asia
Site: Central Research Institute of Electric Power Industry (CRIEPI)
Abiko Site, 1646 Abiko, Abiko-shi
Chiba-ken 270-1194 Japan
http://criepi.denken.or.jp/en/ Date Visited: December 5, 2007
WTEC Attendees: L. Petzold (report author), A. Arsenlis, C. Cooper, D. Nelson
Hosts: Dr. Hisashi Kato, Staff Director, International Cooperation Planning Group
Dr. Tetsuo Matsumura, Deputy Director, Materials Science Research Laboratory
Dr. T. Iwatsubo, Senior Research Scientist, Energy Engineering Research Laboratory
Dr. H. Kaieda, Senior Research Geophysicist, Civil Engineering Research Laboratory
Dr. N. Soneda, Senior Research Scientist, Materials Science Research Laboratory
Dr. K. Nakashima, Research Scientist, Materials Science Research Laboratory
Dr. N. Hashimoto, Research Scientist, Energy Engineering Research Laboratory
The Central Research Institute of Electric Power Industry (CRIEPI) is a Japanese nonprofit corporation founded in 1951 with a broad mission of “solving global environmental problems while ensuring energy security.” CRIEPI has an annual budget of ¥36.8 billion,1 which comes mostly from the Japanese power industry; the remainder comes from the Japanese government. CRIEPI has 786 employees of whom 656 are classified as researchers. The institute is organized into one center, the Socio-economic Research Center, and 7 research laboratories: Systems Engineering, Nuclear Technology, Civil Engineering, Environmental Science, Electric Power, Energy Engineering, and Materials Science Laboratories. CRIEPI’s permanent research staff members have formal education in electrical, civil, mechanical, chemical, and nuclear engineering as well as in biological, environmental, information, and social sciences. They publish on the order of 1500 papers per year and file on the order of 300 patent applications per year.
The largest simulation efforts at CRIEPI are focused around materials science and climate/weather simulations. The materials science simulations focus on the multiscale modeling of irradiation embrittlement processes, light water reactor pressure vessel steels, and the stability and growth of oxide films on silicon carbide The multiscale aspects of the effort include molecular dynamics simulations of irradiation cascades, copper precipitation in iron, and dislocation irradiation defect reactions; kinetic Monte Carlo simulations of microstructural coarsening during irradiation; and dislocation dynamics simulations of irradiated strength change. The multiscale modeling framework relies on passing relevant information from small length scale simulations to larger length scale simulations, with the final result being a continuum rate model of irradiation damage evolution and corresponding mechanical property change. The oxide stability simulations focus on performing first principles molecular dynamics simulation of interfaces SiO2 and SiC with defects and observing the chemical reactions of oxygen atoms across the interface. Such interface calculations are at the forefront of computational chemistry and required the use of the Earth Simulator to retain stable interfaces and be able to introduce stable point defects at those interfaces.
The CRIEPI researchers, in collaboration with NCAR (National Center for Atmospheric Research) and Earth Simulator staff, led an international team in performing unprecedented global climate simulations on the Earth Simulator that has influenced international policy on global climate change. The effort was sponsored by MEXT, and the porting of the CCSM 3.0 codes to the vector architecture of the Earth Simulator required approximately 10 person-years worth of effort. The result of the effort was the ability to perform multiple 400-year predictions of future temperature and sea level changes with different CO2 emissions profiles. The simulations required about six months of wall clock time on the Earth Simulator for four different scenarios to be completed. A challenging high-resolution (a grid spacing of 10 km) ocean simulation was also conducted on the Earth Simulator, and eddy-resolved ocean currents were successfully simulated. Since then, the team at CRIEPI has begun performing regional climate simulations of East Asia to assess the regional impacts of global warming, high-resolution weather forecasting simulations, and global models of ocean currents. Several highlights included simulations of a typhoon making landfall in Japan, and predictions of local flooding in areas surrounding Japan’s energy infrastructure that required higher resolution than the Japanese Meteorological Agency typically performs.
CRIEPI researchers provide simulation analysis and tools to the Japanese electric power companies in other areas as well. Examples include CRIEPI’s Power system Analysis Tool (CPAT) for electric grid stability, CFD analyses of combustion in coal-fired power plants, CFD analysis of transmission line vibration, and seismic analysis of oil tanks. For the most part, CRIEPI employs commercial software in its routine simulations and outsources code development to tailor the commercial tools for problems of interest. However, the simulation tools for the high-end materials science and climate simulations are developed in-house or with a small set of international collaborators. CRIEPI management considers the codes developed in-house to be proprietary; after establishing the proprietary rights the institution shares its codes with the scientific community.
The researchers at CRIEPI working in the materials science, combustion, and climate areas are proposing ambitious simulation projects. In the nuclear materials arena, researchers are interested in simulating the process of stress-corrosion cracking of pressure vessel steels and cladding materials using a multiscale modeling framework. The challenging aspect of this endeavor is that it adds the complexity of surface chemistry and interfacial defects to simulations of mechanical behavior in irradiation environments. In the combustion arena, researchers are interested in predicting the combustion of new fuels and helping to design next-generation power plants. In the climate arena, researchers are interested in increasing the spatial resolution of the ocean current simulations to 0.1 degree such that oceanic eddies could be resolved, in adding complexity and explicit degrees of freedom in their climate models, and in developing a full “Earth system model” that simultaneously models the evolution of the atmosphere, ocean currents, and land masses.
An equally impressive experimental program was presented in the areas of nuclear materials and combustion science. CRIEPI has invested in state-of-the-art experimental hardware that is able to probe material microstructure at the same level as its kinetic Monte Carlo simulations of irradiation damage evolution. Its experimental and simulations efforts are well coordinated, and there appears to be strong feedback between the two. Validation of the kinetic Monte Carlo simulations allows CRIEPI researchers to make credible predictions of material behavior at engineering length scales with knowledge of the proper mechanisms of microstructural evolution. In their combustion experiments, CRIEPI researchers have built small test furnaces to validate the results and CFD simulations, and it appears that the simulators are now driving the experimentalists to make finer measurements of processes than they have had in the past. SBES at CRIEPI appears to have changed the manner in which experimental data is assessed.
In its hiring practices, CRIEPI tends to hire domain specialists when seeking new people for the institution’s modeling and simulation activities and does not exclusively seek out candidates with experience in computer science disciplines. The research staff preferred to hire domain specialists and train them to perform computer simulations rather than hire computer scientists and train them in the technical disciplines that require simulation support. There was an opinion shared by the staff that the universities were properly preparing students to perform the sophisticated cross-cutting research needed at industrial institutions.
CREIPI has an impressive array of simulation and experimental capabilities to address the most challenging problems facing the Japanese power industry. Its high-end simulations capabilities in materials science and global climate change are on par with the premier simulation institutions around the world. The institution’s mission focus enables its researchers to build strongly coupled simulation and experimental campaigns enabling them to validate their simulation results and build confidence in predictions. The close coordination of simulation and experiment should serve as an example to the rest of the world..
Site:Computational Materials Science Center (CMSC)
National Institute for Materials Science (NIMS)
Tsukuba, Japan Date Visited: December 7, 2007
WTEC Attendees: S. Glotzer (report author), M. Head-Gordon, S. Kim, J. Warren, P. Westmoreland
Hosts: Dr. Masaki Kitagawa, Vice President, NIMS
Dr. Shigeru Okamura, International Affairs Office, NIMS
The National Institute for Materials Science was formed in 2001 by combining the National Research Institute for Metals (founded in 1956) and the National Institute for Research in Inorganic Materials (founded in 1966). The NIMS mission is (1) fundamental research and generic/infrastructural technology R&D; (2) popularization of research results and promotion of related activity; (3) common use of NIMS facilities and equipment; and (4) training of researchers and engineers and improvement of the quality of these human resources. NIMS is somewhat similar in scope to the Materials Science and Engineering Laboratory at the National Institute of Standards and Technology in the US. Their research is of the highest quality and impact. The have 124 MOUs with countries around the world, and 12 sister institutes in 7 countries. They participate in 10 graduate schools internationally, and collaborate extensively with industry. NIMS ranked 6th in the world in # of citations in materials. Six major research fields for NIMS from 2006-2010 are:
Nanotechnology-driven advanced materials research
Nanotech common key technologies
Synthesis and control of novel nanomaterials
Nanotech driven materials research for information technology
Nanotech driven materials research for biotechnology
Advanced materials research for social needs
Materials research for the environment and energy
Materials research for reliability and safety
The Computational Materials Science Center (CMSC) within NIMS aims to develop advanced simulation technologies for nanoscale materials with innovative properties and explore design rules for novel properties and functions. The Center analyzes and predicts properties of nanomaterials and aims to clarify structure-property relationships. Tools include various advanced simulation techniques such as large-scale first-principles simulations, function analysis simulations, strong coupling models, phase-field modeling, and multiscale modeling and simulations.
The director of the CMSC is Dr. Takahisa Ohno. The Center is comprised of four groups, and totals roughly 20 permanent research staff and 20 postdocs and graduate students:
First Principles Simulation Group I – One group leader, six researchers and engineers, and 12 research fellows, and two administrative staff. Dr. Ohno is also the group leader of the First Principles Simulation Group I.
First Principles Simulation Group II - – One group leader, five researchers and engineers, four research fellows, and one office assistant.
Strong Coupling Modeling Group – One group leader, three researchers and engineers, three research fellows, and one office assistant.
Particle Simulation and Thermodynamics Group – One group leader, six researchers and engineers, four research fellows, and one office assistant.
The visiting panel heard an overview of NIMS and CMSC research from the Dr. Kitagawa, VP of NIMS, and Dr. Sasaki, Group Leader in CMSC.
SBE&S methods within the Center span multiple scales. First principles simulations include order-N DFT simulations, hybrid methods, and TDDFT. Nanoscale functional analysis includes electron transfer, spin transport, and diffusion quantum Monte Carlo. Strong coupling modeling treats quantum transport, many-body effects, and thermal fluctuations. Multiscale modeling and simulation combines molecular dynamics, Monte Carlo, and phase-field methods. The phase field work is particularly highly regarded within NIMS. Number 2 of the top 3 materials infrastructure accomplishments at NIMs is the structural materials database (see Software below).
The panel was impressed with the quantity and quality of the SBE&S research we heard about. Overall, the computational research is highly sophisticated and on par with the best work in the US and Europe. Current projects within each of the four groups of the CMSC include:
First-principles simulation group I:
(a)Development of Calculation Method for Large-scale DFT Simulation
(b)Investigation of Electron Transport Properties of Nanostructures
(c)First Principles Simulation of Redox Reactions
(d)Control of the States of Materials
First-principles simulation group II:
(a)Electron Correlation Effects on Materials Properties
(b)Prediction of Properties of Materials
(c)Response of Materials (including magnetic, superconducting and optical materials)
Strong-coupling modeling group
(d)Crystal-Orbit-Spin Coupling in Novel Cuprates and Materials Design
(e)Investigation on Phase Transition and Dynamics of Vortices in Superconductors
(f)Research on Quantum Transport Phenomena
Particle simulation and thermodynamics group
(g)Prediction of Microstructure Evolution by Phase-Field Method
(h)Reaction Diffusion Behavior between Ni-Al-Ir Alloy
(i)Research on Nanostructure of Materials by MD Simulations
The main supercomputer is the Hitachi SR11000/62 with 992 processors and 2 Tb storage. A new major acquisition from a Japanese computer vendor is planned for 2008. They are not currently addressing multicore computing future architectures.
Many of the codes used by the CMSC are developed in house. The CMSC has developed several impressive SBE&S software platforms now used by others. These include:
A thermodynamics database and simulation software developed in collaboration with AIST, Tohoku Univ., Kyushu Inst. Tech., and InterScience Ltd. The software company name is Materials Design Tech. CO. LTD, a venture company from NIMS.
MatEX: A Materials Design Platform (http://matex.nims.go.jp/). Demonstration programs for the phase-field modeling of microstructure evolution in engineering materials on the web.
Nanoscale Device Simulation Software (http://www.rss21.iis.u-tokyo.ac.jp). They offer services to predict and design the materials properties and functions for the next-generation semiconductor nano-devices on the web.
The materials database involves an exchange of personnel with de la Rubia’s group at LLNL.
During the talks there was good opportunity for discussion of both scientific issues directly related to the content of the talks, and also more general issues surrounding simulation-based engineering and science. Some of the additional issues that arose included:
Education and HR in SBE&S: Potential postdocs and research staff with appropriate training in the development of algorithms and programs are hard to recruit. The main cause of this issue is that too many graduate students are being trained primarily to run existing codes to solve applied problems rather than learning the skills necessary to create a new application. NIMS recently created the International Center for Young Scientists (ICYS) to become a more attractive workplace for foreign young scientists. The concept is based upon (1) international: English as working language; (2) independent: autonomous research; (3) interdisciplinary: a fusion of different cultures and fields; and (4) innovative: strategic research. Benefits to participants in this program include a high salary, research grant support (5 M yen/year), and ideal research environment with cutting edge facilities and single occupancy cubicles. In five years, CMSC hired 10 simulators. Most of their backgrounds are in materials science, some physics. Most know how to run codes, but not how to innovate and develop new codes.
Appreciation of SBE&S: SBE&S activities at NIMS started 15 years ago, and CMSC started in 2001 with support from the Director-General. With new emphasis on nanoscience, experimentalists are becoming more interested in collaborating with simulators. Experimentalists now seek out input from simulators, unlike a several years ago. Simulation now having more and more impact on real problems.
Interaction with industry: No current interactions with industry.