1. The Handling of Crystallographic Data
Brian McMahon, International Union of Crystallography, England

The Crystallographic Information File (CIF) was commissioned by the International Union of Crystallography in the late 1980s to provide a common exchange format for diffraction data and the structural models derived therefrom. It specifically addressed the requirements of an information exchange mechanism that would be portable, durable, extensible and easy to manipulate, and has won widespread acceptance as a community standard. Nowadays, CIFs are created by diffractometer software, imported and exported from structure solution, refinement and visualisation programs, and used as an electronic submission format for some structural science journals.

CIF employs simple tag-value associations in a plain ASCII file, where the meanings of the tags are stored in external reference files known as data dictionaries. These dictionaries are machine-readable (in fact conforming to the same format), and provide not only a human-readable definition of the meaning of a tag, but also machine-parsable directives specifying the type and range of permitted values, contextual validity (whether an item may appear once only or multiple times) and relationships between different items. In many ways this is similar to the separation between document instances and their structural descriptions (document type definitions or DTDs) in XML, the extensible markup language that is increasingly used for document and data handling applications. However, while many existing XML DTDs describe rather general aspects of document structure, the tags defined in CIF dictionaries detail very specific pieces of information, and leave no room for ambiguity as these items are read into and written out from a variety of software applications.

Recognised tags in CIF include not only subject-specific items (e.g. the edge lengths of a crystal unit cell) but also general tags describing the creator of the file (including address and email), its revision history, related literature citations, and general textual commentary, either for formal publication or as part of a laboratory notebook record. The objective is to capture in a single file the raw experimental data, all relevant experimental conditions, and details of subsequent processing, interpretation and comment. From a complete CIF, specialist databases harvest the material they require. While such a database might be unable to store the entire content of the source file, the IUCr encourages databases to retain deposit copies of the source or to provide links from database records to the source (for example as a supplement to a published journal article).

The richness of the tag definitions also allows automated validation of the results reported in a CIF by checking their internal consistency. At present validation software is built by hand from the published descriptions of data tags, but experiments are in hand to express the relationship between numeric tags in a fully machine-readable and executable formulation. While the CIF format is unique to crystallography (and a small number of related disciplines) it has much to contribute towards the design of similar data-handling mechanisms in other formats.

2. Development of KDB (Korea Thermophysical Properties Databank) and Proper
Use of Data and Models in Computer Aided Process Engineering Applications
Jeong Won Kang, CAPEC, Technical University of Denmark, Denmark
Rafiqul Gani, Technical University of Denmark, Denmark
Chul Soo Lee, Korea University, Korea
Ki-Pung Yoo, Sogang University, Korea

The physical property data, equilibrium data and prediction models are essential parts of process synthesis, design, optimization and operation. Although efforts to collect and organize such data and models have been performed for decades, the demand for data models and their proper and efficient use are still growing. With the financial support of MOCIE (Ministry of Commerce, Industry and Energy) of Korea, four universities have collaborated to develop a thermophysical properties databank and enhance their capacity on experimental production. The databank (KDB) contains about 4000 pure components (hydrocarbons, polymers and electrolytes) and 5000 equilibrium data sets. Most of the data were collected along with their accuracy of measurements and/or experimental uncertainties. The data can be searched by a stand-alone program or via internet. This presentation will discuss current status and features of KDB.

In process engineering applications, selecting proper data, selecting proper model, regression of the model parameter and their proper uses are the most important aspect. CAPEC( Computer Aided Process Engineering Center, Technical University of Denmark) has been developing programs to help the proper use of thermodynamic properties data and prediction models for years. A stepwise procedure to select data sets from property databases such as KDB and CAPEC-DB , generating problem specific parameters and their proper use through appropriate property models in process engineering problems has been developed in CAPEC. The presentation will also highlight the application of property model and data in specific process engineering problems.

 

3. Reliability of Uncertainty Assignments in Generating Recommended Data from a Large Set of Experimental Physicochemical Data
Qian Dong, National Institute of Standards and Technology, Boulder, CO, USA

Experimental (raw) physicochemical property data are the fundamental building blocks for generating recommended data and for developing data prediction methods. The preparation of recommended data requires a well-designed raw data repository with complete supporting information (metadata) and reliable uncertainty assessments, a series of processes involving data normalization, standardization, and statistical analysis, as well as anomaly identification and rectification. Since there are considerable duplicate measurements in a large data collection, uncertainty assessments become a key factor in selecting high quality data among related data sets. While other information in the database can help with the selection, the uncertainty estimates provide the most important information concerning the quality of property data. This presentation will focus on the assignment and assessment of uncertainty with a large set of experimental physicochemical property data as well as the impact of uncertainty assessments on generating recommended data.

Uncertainties represent a crucial data quality attribute. They are stored in the form of a numerical value, which is interpreted as a bias for the associated property value. The addition and subtraction of this bias from the property defines a range of values. Without uncertainties, numerical property values cannot be evaluated, while inappropriate uncertainties can also be misleading. In assessing uncertainty all potential sources of errors are propagated into the uncertainty of the property. In this process, complete information on measurement techniques, sample purity, uncertainty assessment by the investigator, and investigator's experience/records, etc. is essential in establishing uncertainties by database professionals.

Reliable provision of uncertainties for property values in databases establishes the basis for determination of recommended values. However, the process of arriving at an appropriate judgment on uncertainties is rather complex. Correct assignment of uncertainty requires highly knowledgeable and skilled data professionals, and furthermore, includes a subjective component. A large-scale data collection such as TRC SOURCE makes this sophisticated task even more demanding. A recent statistical analysis on critical constants and their uncertainties assigned in TRC SOURCE reflected the difficulty in assigning reliable uncertainties and also revealed a decisive effect of uncertainties on generating recommended values. Based on this study, a computer algorithm has been developed at NIST/TRC to systematically evaluate uncertainty assessments.

4. Dortmund Data Bank (DDB)- Status, Accessibility and Future Plans
Jürgen Rarey and Jürgen Gmehling, University of Oldenburg, Germany

With a view to the synthesis and design of separation processes, fitting and critical examination of model parameters used for process simulation and the development of group contribution methods 1973 a computerized data bank for phase equilibrium data was started at the University of Dortmund. While at the beginning mainly VLE data for non-electrolyte mixtures (T_b > 0 °C) were considered, later on also VLE (including compounds with T_b < 0 °C), LLE, h^E, γ^∞, azeotropic, c_P^E, SLE, v^E, adsorption equilibrium, ... data for non-electrolyte and electrolyte systems as well as pure component properties were stored in a computer readable form. This data bank (Dortmund Data Bank (DDB)) now contains nearly all worldwide available phase equilibrium data, excess properties and pure component properties.

To use the full potential of this comprehensive compilation a powerful software package was developed by DDBST GmbH (www.ddbst.de) for verifying, storing, handling and processing the various pure component and mixture data. Programs for the correlation and prediction of pure component properties, phase equilibria, excess properties as well as graphical data representation were also included.

Together with the data from the Dortmund Data bank these programs allow to analyze the real mixture behavior of a system of interest and to fit reliable model parameters (g^E-models, equations of state, group contribution methods) for the synthesis and design of chemical processes on the basis of the most actual experimental data and estimation methods.

The talk will give an overview on the development, structure and contents of the DDB and will highlight certain aspects of the accessibility and use of thermophysical data in the Internet age. Future plans concerning the development of the DDB and the software package DDBSP will be discussed.

Track I-D-1: Data On Gas Hydrates Chair: Fedor Kuznetsov, Inst. Inorg. Chem., Novosibirsk, Russia The session will be devoted to a discussion of the status of data on gas hydrates. It is of great interest now in many countries to find reliable and economically viable ways to use the huge resources stored in nature in the form of solid gas hydrates in permafrost areas and at the bottom of the ocean. Recovery of gas from these deposits is an extremely complicated undertaking. Exploration of the deposits, development of technologies for gas recovery, conditioning and transportation, prevention of ecological hazards – all of these problems require a great variety of different data. The session will include presentations on general problems of data collection and management as well as information on data activity in this field in different countries.

1. Gas hydrates in Siberian geological structures
Albert D. Duchkov, Institute of Geophysics SB RAS, Novosibirsk, Russia

Results of prospecting of gas hydrates accumulations in continental regions of Siberia are discussed.

In Russia, the problem of existence of gas hydrates (GH) deposits is usually discussed in the context of hydrate saturation of the Cenomanian gas pool at the Messoyaha deposit in east northern part of West Siberia. However GH were never directly observed at the Messoyaha gas deposit during 40 years of investigations. One more producing horizon has been recognized in the same geological area. This horizon is related to the Cazsalin Layer of Turonian-Coniacian age, lying above the Cenomanian deposits and having more favorable PT- conditions for hydrates formation. Analysis of specific features of geologic structure, temperature regime of the section, gas composition, mineralization of formation waters, logging data, seismic prospecting materials, and sampling suggests that gas hydrates can exist in the Cazsalin Layer of the East Messoyakha deposit. One of the possible directions of further study of genesis of natural gas hydrates and estimation of the effect of gas hydrates processes on the structure of gas deposits and gas resources is study of the hydrocarbons accumulated in the Cazsalin Layer of the East Messoyakha deposit with sampling of core by a sealed thermostatically controlled corer.

The GH accumulations were found in Lake Baikal (East Siberia). Multichannel seismic studies, performed during 1989 and 1992, have revealed in Baikal the "bottom-simulating reflector" (BSR), which gives an exact evidence of existence of the lower boundary of the GH layer. It has been established that gas hydrates are distributed in South and Central parts of Lake everywhere in places where the water depth is more than 500 meters. Four types of tectonic influence were revealed: 1) modern faults shift the BSR as they do it with usual seismic boundaries; 2) older faults shift normal reflectors, the BSR has no shifts; 3) modern faults form zones, where the BSR is destroyed; 4) the processes that proceed within older faults situated closely to the base of hydrated layer leads to undulations of the BSR. The depth of lower boundary of the GH layer in Baikal ranged from 35 to 450 m. The GH presence in the Lake Baikal sediments has been confirmed by underwater borehole BDP-97 and special geological investigations. The GH accumulations were found at the surface of bottom and in sands at the depth interval 121-161 m below bottom.

• Properties of hydrates and hydrate-saturated media
• Conditions of formation and dissociation of hydrates in porous media
• Effective technologies for production of gas from offshore and permafrost hydrates
• Optimum conditions for storage and transportation of gas in hydrate state
• Influence of gas hydrates on the Earth environment

 

3. Data on kinetics and thermodynamics of gas hydrates, application to calculations of phase formation
John A. Ripmeester, SIMS, NSC, Canada

Experimental data on gas hydrates are being produced at a rapid rate, and arise from laboratory studies, field studies and industrial laboratories, each working independently. The international hydrate community has an increasing need to access reliable data on the structural and physicochemical properties on hydrates in an efficient way. The creation of an information system covering all issues relating to hydrates is essential, as this is necessary for the prediction of hydrate occurrences, both in natural and industrial environments and the control of hydrate formation and decomposition. Ultimately this will affect our ability to predict the existence of hydrate-related hazards, to judge the potential for hydrates to contribute to the global energy supply as well as their possible influence on climate change.

 

4. Gas Hydrates Management Program at GTI
A. Sivaraman, Gas Technology Institute, USA

Gas hydrates are an impediment to gas flow as well as a potential energy resource. When they form inside pipelines, hydrates can slow or completely block gas flow, a significant problem for producers striving to move gas from offshore wells to onshore processing facilities. Producers, gas storage, transmission companies spend millions of dollars each year on hydrate inhibitors and other actions to help prevent hydrate formation, trying to balance cost, environmental impact, efficiency and safety. Better understanding of the mechanisms that trigger hydrate formation and dissociation could lead to creation of more effective hydrate inhibitors.

The U.S. Department of Energy, Gas Research Institute (Currently GTI) and U.S Geological Survey have documented the presence of hydrates in artic Alaska, off the U.S. Atlantic and Pacific coasts, as well as in the Gulf of Mexico and the hydrate deposits contain as much as 320,000 Tcf of natural gas compared to the current consumption of 22.5 Tcf per year in United States. Various joint industry programs are focused in drilling and producing gas from gas hydrate fields in deep waters off the coast in US and Japan.

GTI is the premier, industry-led natural gas research and development organization in the United States, dedicated to meet current and future energy and environmental challenges. At its facilities near Chicago, Illinois, GTI has assembled state-of-the art laboratories (Laser Imaging, Acoustics and Calorimetry) operated by an expert research team that is uniquely equipped to investigate the mechanism of formation and dissociation of gas hydrates; the impact of drilling fluids, the low dosage inhibitors and anti agglomerents on the hydrates. Recent results from the facility are presented.

 

5. Computer Modeling of the Properties of Gas Hydrates - The state-of-the-art
John S. Tse, Steacie, Institute for Molecular Sciences, National Research Council of Canada

Various theoretical techniques for the modelling of the physical, thermodynamics and electronic properties of gas hydrates will be reviewed. Selected examples from recent work of the author's group will be presented. Emphasis will be placed on the prediction of the dynamic properties, occupancy, formation and dissociation mechanism of gas hydrates. Perspective on using advanced simulation method for the prediction of phase equilibria will be discussed.

 

6. Natural Gas Hydrates Studies in China
Shengbo Chen and Guangmeng Guo, Institute of Geography Sciences and Natural Resources Research, China

Natural gas hydrates studies are very important. The CODATA Task Groups on Data on Natural Gas Hydrates was newly approved in October 2000. In China, gas hydrates is a potential field for studying and exploring. The area of permafrost regions accounts for 10% of the world permafrost, especially in the mid-latitude and high-altitude mountainous regions in Qinghai-Tibet Plateau. The oil-gas resources have been confirmed by exploring in the north of Tibet Plateau. It is made clear that methane emissions and carbon dioxide uptake by observation in Qinghai-Tibet Plateau. These evidences show volumes of gas hydrate may be exist. In addition, extensive sea and long shoreline make it hopeful that began to study and explore gas hydrates. In China offshore seas, mainly in South China Sea and East China Sea, obvious signs of hydrates have been distinguished in seismic reflection profile, and high temperature of seawater and high ratio of methane in fluids can be observed. All these signs and observations indicate that it is completely possible there exists a large amount of gas hydrates in China offshore seas.

In 1990, the first experimental forming of gas hydrate was finished by composing methane and water vapor in China. Subsequently, more and more university, institute and corp. involve in gas hydrates studies, including thermodynamics of hydrate formation/decomposition, seismic observation and geochemical analysis. For example, the 9 bottom simulating reflection (BSR) was found in South China Sea, and methane contents analysis by collecting sample in East China Sea have been carried out. The information of new Earth Observation System (EOS), including EOS-MODIS and EOS-MOPITT is being applied to exploring gas hydrate in Qinghai-Tibet Plateau. The land surface temperature information in the permafrost can be retrieved by the infrared data of EOS-MODIS, and the methane emissions and carbon dioxide uptake can also gained easily by EOS-MOPITT. Actually, the high temperature of sea surface by the infrared data retrieving is consistent with the distribution of high ratio of methane in fluids in East China Sea, which proved it is possible by using EOS information.

 

7. State of CODATA project on information system on Gas Hydrates
Fedor A. Kuznetsov, Institute of Inorganic chemistry SB RAS, Chairman of CODATA Task Group, Russia.

Previous CODATA general assembly approved establishment of task group on Gas Hydrates data. Most authoritative specialists in field of gas hydrates were invited to be members of the group. They in total represent all major field of science and technology related to gas hydrates and most of the countries, were gas hydrates studies attract significant attention. The group has developed a concept and general recommendations on the structure of the system and requirements of data.

The system thought of as a network of independent data groups, which make their own data bases in field of their expertise. What makes this network a distributed information system is set of requirements for data management accepted by all the participants. The planned system will cover information from the following areas of science and application in relation to gas hydrates:

By now more then a hundred groups in different countries identified by now as prospective participants of creation of the system.

Present state of the system and plans for future will be reported.

Track III-C-1: Materials Databases Chair: Huang Xinyue

1. Molten Salt Database Project: Building Information and Predicting Properties
Marcelle Gaune-Escard, Ecole Polytechnique, France

The genesis of the Molten Salt Database, realized as early as 1967 with the publication of the Molten Salt Handbook by George Janz is as relevant today as it was over 30 years ago. New high-tech applications of molten salts have emerged and the need for data is crucial for the development of new processes (pyrochemical reprocessing of nuclear fuel, nuclear reactors of new generation, elaboration of new materials, new environment-friendy energetic sorces, …).

Building a world-class critically, evaluated database is a difficult and complex process, involving considerable time and money. Ultimately, the success of the project depends on positive interactions between a diverse group of people - support staff to identify and collect relevant literature, scientists to extract and evaluate the data, database experts to design and build the necessary data architecture and interfaces, database reviewers to ensure that the database is of the highest quality, and marketing staff to ensure the widest dissemination of the database. The advent of the World Wide Web (WWW) has provided another exciting component to this paradigm - a global database structure that enables direct data deposition and evaluation by the scientific community.

Also the new concepts in engineering data information system are emerging and make it possible to merge people, computers, databases and other resources in ways that were simply never possible before.

Ongoing efforts in this respect will be described with the ultimate goal of building a Virtual Molten Salt Laboratory.

These efforts are made in parallel with our current research activities on molten salts but also in interaction with those other related actions on materials and engineering. For instance, it is also intended to adapt and apply methodologies originally used for other purposes ("human genome") to the field of molten salts., as recently demonstrated for other materials by K. Rajan at RPI, using computational "informatics" tools.

 

2. Development of Knowledge Base System Linked to Material Database
Yoshiyuki Kaji, Japan Atomic Energy Research Institute (JAERI), Japan
Hirokazu Tsuji, Japan Atomic Energy Research Institute (JAERI), Japan
Mitsutane Fujita and Junichi Kinugawa, National Institute for Materials Science, Japan
Kenji Yoshida and Kazuki Shimura, Japan Science and Technology Corporation, Japan
Shinichi Mashiko and Shunichi Miyagawa, Japan Nuclear Cycle Development Institute, Japan
Shuichi Iwata, University of Tokyo, Japan

The distributed material database system named 'Data-Free-Way' has been developed by four organizations (the National Institute for Materials Science, the Japan Atomic Energy Research Institute, the Japan Nuclear Cycle Development Institute, and the Japan Science and Technology Corporation) under a cooperative agreement in order to share fresh and stimulating information as well as accumulated information for the development of advanced nuclear materials, for the design of structural components, etc. In the system retrieved results are expressed as a table and/or a graph.

In order to create additional values of the system, knowledge base system, in which knowledge extracted from the material database is expressed, is planned to be developed for more effective utilization of Data-Free-Way. A standard type retrieval screen is prepared for users' convenience in Data-Free-Way. If typical retrieved results through the standard type retrieval screen are available, users do not need to retrieve the data under the same condition. Moreover, if the meaning of the retrieved results and the analyzed results are stored as knowledge, the system becomes more beneficial for many users. As the first step of the knowledge base development program, knowledge notes have been made where typical retrieved results through the standard type retrieval screen and the meaning of the retrieved results are described by each organization. XML (eXtensible Markup Language) has been adopted as the description method of the retrieved results and the meaning of them. One knowledge note described with XML is stored as one knowledge which composes the knowledge base. Knowledge notes can be made at each stage of the data retrieval, the display of the retrieved results, or the graph making. A set condition at each stage can be reproduced from the knowledge note. Storing knowledge obtained as retrieved results are described with XML. And a knowledge note can be displayed using XSL (eXtensible Style Language). Since this knowledge note is described with XML, the user can easily convert the display form of the table and the graph into the data format which the user usually uses. Moreover, additional information to the retrieved numerical values such as a unit can be easily conveyed.

This paper will describe the current status of Data-Free-Way and the description method of knowledge extracted from the material database with XML.

3. Activity on Materials Databases in the Society of Materials Science, Japan
Tatsuo Sakai, Ritsumeikan University, Japan
Izuru Nishikawa, Osaka University, Japan
Atsushi Sugeta, Osaka University, Japan
Toshio Shuto, Mitsubishi Research Institute Inc., Japan
Masao Sakane, Ritsumeikan University, Japan
Tatsuo Inoue, Kyoto University Sakyo-ku, Japan

The data book, consisting of Vols.1, 2 and 3, was published in 1982 by the Society of Materials Science, Japan (JSMS). Volumes 1 and 2 contained numerical data of fatigue strength of metallic materials, and Vol.3 contained graphic presentations of the data. All the data were compiled as a machine-readable database and the database was opened to use in the research and engineering applications. Furthermore, after collecting additional new data, the serial data book was also published in Vols.4 and 5 in 1992 from the same society, and these data were also compiled as the database. The CGS unit system was used in Vols.1, 2 and 3, but the SI unit system was employed in Vols.4 and 5.

In order to facilitate the useful application, both data books were combined with each other as a fully revised version, and a new data book of three volumes was published by Elsevier Science B. V. and JSMS in 1996. The database was similarly revised as a new version and it was circulated as several types of medias such as Floppy Disc, DAT-Tape domestically in Japan. These databases have been widely used in the engineering applications in Japan.

In accordance with the progress of information technology, requirements to the materials database were markedly increased in the last decade. Thus, JSMS had organized some new projects to construct two other kinds of databases in the area of materials science. The first one is the database on tensile and low-cycle fatigue properties of solders. The objective materials are Sn-37Pb and Sn-3.5Ag solders, respectively. The second one is the database on the material characteristics such as stress-strain curves and temperature dependence of heat conductivity, specific heat, elongation and Young’s modulus. These databases were circulated as CD-ROM domestically in Japan.

In the present conference, the historical scope of the database construction in JSMS and their contents are introduced together with some examples of their engineering applications performed by some research groups in JSMS. Making reference to discussions in the present conference, the authors are looking for the effective method to circulate the JSMS databases in the worldwide scale.

 

4. Role of MITS-NIMS to Development of Materials database
Y. Xu, J. Kinugawa and K. Yagi, National Institute for Materials Science (NIMS), Japan

Material Information Technology Station (MITS) of National Institute for Materials Science (NIMS), established in October 2001, is aimed to be a worldwide information center for materials science and engineering.

Our main activities include fact-data producing and publication, literature data acquisition, and database production. We have been continuing experiments of metal creep and fatigue for 35 years, and the data are published and distributed as NIMS Data Sheets. Besides, from this year, we start literature data acquisition on materials' structure and properties. Both of the fact-data and literature data are stored and managed as databases. We are constructing more than 10 material databases, which include polymers, metals and alloys, nuclear materials, super conducting materials, etc. Online services of these databases will be available from next April.

Being aware that a simple system with only data retrieving function can not provide enough information for material research and industrial activities, in which not only data, but also data related knowledge, and decision support function are needed, we have started several new research and development projects aiming to construct intelligent material information systems with data integration, data analysis and decision support functions.

One of our projects is to develop a material risk information platform. Basing upon material property databases, material life prediction theory, and accident information databases, this platform will provide users with material risk knowledge as well as fact data, for the purpose of safe use and correct selection of materials used for high risk equipment, for example, a power plant.

Another system under construction is a decision support system for composite material design - a composite design and property prediction system. With this system, a virtual composite can be composed with optional structure and component materials. Then some basic properties such as thermal conductivity of the composite can be evaluated according to its constitution and the properties of constituents that stored in the databases.

Track III-D-1: Physical/Chemical Data Issues Chair: Marcelle Gaune-Escard, Ecole Polytechnique, France

1. Thermodynamic Properties and Equations of State at High Energy Densities
V. E. Fortov, Institute for High Energy Densities, Russian Academy of Sciences, Moscow, Russia

During last century the range of thermodynamic parameters was greatly broadened because of rapid development of technologies. Thermodynamic properties of matter at high pressures and temperatures are very important for fundamental researches in the fields of nuclear physics, astrophysics, thermodynamics of dense plasma. A number of applications such as nuclear fusion, thermonuclear synthesis, creation of new types of weapon, comet and meteorite hazard etc. requires knowledge of experimental data in a wide region of parameters.

Traditional way of studying of thermodynamic properties of substances at high temperatures and pressures is shock-wave experiments. During last 50 years there have been published about 15000 experimental points on shock compression, adiabatic expansion and measurements of sound velocity in shock compressed matter. These data are used to determine the numerical coefficients of general functional dependencies found from theoretical considerations in semiempirical equations of state (EOS). In this work presented are different semiempirical EOS models which are used for generalization of experimental and calculated data: from simple caloric models for organic compounds and polymer materials to complex multiphase equations of state for metals. These EOS models are valid in a wide range of phase diagram and describe experimental data with good accuracy. These models are also included into the database on shock-wave experiments with public access. The database allows one to perform calculations of EOS for large amount of substances and compare the results with experimental data in graphic form via Internet by address: http://teos.ficp.ac.ru/rusbank/.

2. Internet Chemical Directory ChIN Helps Access to Variety of Chemical - Databases on Internet
Li Xiaoxia, Institute of Process Engineering (formerly Institute of Chemical Metallurgy), Chinese Academy of Sciences, China
Li Guo, Hongwei Yang, Fengguang Nie, Zhangyuan Yang & Zhihong Xu

ChIN is a comprehensive directory of Internet chemical resources on Internet and is constructed on an information base approach other than a merely collection of chemistry related links. The daily maintenance of ChIN is done with the aid of ChIN-Manager, a specific tool based on database for maintaining flexible categories, for creating resource summary pages based on different data models. ChIN has been widely recognized in China. ChIN has been summarized as a site with a huge set of evaluated resources for chemists of all disciplines by ChemDex Plus of ChemWeb.com. ChIN has been considered also as the best Internet chemistry resources index in China by ChemDex of University of Sheffield, which is a well known web directory of chemistry in the world.

As chemical databases are the basic daily tools for chemists to get information, chemical database category is one of the most important categories in ChIN. More than 300 databases have been indexed in ChIN, covering varies databases such as bibliographic databases, chemical reactions databases, chemical catalogs, databases for material safety, databases for physical properties, databases for spectroscopy, materials databases, environmental databases, chemists phone books and so on. There is also a subcategory for selected news on the progress of major commercial chemical databases. There is a summary page for each database indexed in ChIN, summary pages for related databases that may classified into different subcategories are cross linked. Up to now, among the chemical databases indexed within ChIN, more than 70 databases can be freely accessible and over 20 databases provide free searching. The total successful requests to ChIN is over 2.5 millions since 1998 and about half are from oversea visits. About 15% requests go to indexing pages of chemical databases in ChIN.

References
1. ChIN Page, http://www.chinweb.com.cn/, the former URL is http://chin.icm.ac.cn/
2. Xiaoxia Li, Li Guo, Suhua Huang, Zonghong Liu, Zhangyuan Yang, “Database Approach in Indexing Internet Chemical Resources”, World Chemistry Congress, Chemistry by Computer,
OB28, Brisbane, 1 - 6 July 2001

 

3. Graph-Theoretical Concepts and Physicochemical Data
Lionello Pogliani, Dipartimento di Chimica, Università della Calabria, Italy

The molecular polarizabilities of fifty-four organic derivatives have been optimally modeled, the induced dipole moment of another set of sixty-eight organic compounds, have, instead, been less optimally modeled. The modeling was performed by the aid of particular descriptors that have been derived by the aid of graph theoretical concepts. Till recently the starting point of these modeling strategies was the hydrogen-suppressed chemical graph and pseudograph of a molecule, which for second row atoms worked quite fine. For each type of graph or pseudograph an adjacency matrix can be written. Actually the pseudograph matrix is enough to represent mathematically either a graph or a pseudograph of a hydrogen-suppressed molecule, as it encodes not only information on single connections but also on multiple connections and self-connections, which mimic multiple bonds and non-bonding electrons. From these matrices specific theoretical graph-structural invariants can be derived, among which the molecular connectivity indices and pseudoindices. For molecules with higher-row atoms, i.e., atoms with n>2, the graph representation alone was insufficient to derive a specific invariant and use was done of atomic concepts, as there was no other way to encode the contribution of the inner-core electrons of higher-row atoms. Recently, and for the first time, inner-core electrons have been successfully 'graph' encoded by the aid of complete odd-graphs, Kp, and of the corresponding adjacency Kp-pseudograph matrix. The use of complete odd-graphs to derive graph-theoretical invariants allowed an optimal modeling of the molecular polarizabilities and a not too bad modeling of the induced dipole moment of organic derivatives of better or similar quality than the modeling achieved by MM3 calculations. Other types of 'non-pure' graph-theoretical invariants achieved less satisfactory modelings.

 

4. Progress in the Development of Combustion Kinetics Databases for Liquid Fuels
Wing Tsang, National Institute of Standards and Technology, USA

This presentation is concerned with the development of a database for the simulation of gas phase combustion. In recent years simulation have become an important tool in technology. The key for effective simulations is a reliable database of information that form the essential inputs. In the area of combustion, the complexity of the process has made necessary the building of large databases. This has been hindered by the fact that gas kinetics, the disciplinary field responsible for generating the database is still a research area. Thus there has been a need for constant upgrading. Even more serious is that most combustion is carried out with liquid fuels which are complex mixtures of intermediate sized hydrocarbons. Normal alkanes are important components and they may contain ten or more carbons.

There has been considerable recent work on the oxidation of various fuels. For one or two carbon fuels there is a state of the art database GRIMECH. There are also databases that describe the formation of PAHs and soot. A complete database should contain sufficient information that will cover the oxidation and pyrolysis reactions leading to soot formation. It should also start with some important components of liquid fuel. We have now started work in this direction using heptane as our prototypical liquid fuel. The aim is to develop the kinetics sufficiently so that they can be interfaced with the existing databases mentioned earlier.

The need is for a database specifying the thermal cracking reactions of the fuel. These can be classified as bond breaking of the fuel, decomposition of the radicals formed from bond breaking and/or radical attack, decomposition of olefins, the first stable product from radical decomposition, and finally the decomposition of the olefinic radicals. This will define the nature of the competition between oxidation and cracking and the small unsaturated species that are starting point for PAH formation and soot models.

Among the four classes of reactions, processes beginning with stable compounds are in satisfactory conditions in the sense of the availability of experimental data or methods for estimation. The technically difficult problem is the quantitative specification of the decomposition modes of the radicals. This is due to the fact that larger alkyl radicals can also isomerize. Thus for many cases it is necessary to consider the decomposition modes of all the isomerization products simultaneously. Furthermore due to their low reaction thresholds energy transfer effects must be considered. This means that reaction rates are pressure as well as temperature dependent. We will describe how this problem has been solved in the C5-C7 radicals. Finally we show how the present results lay the basis for the extension of the database too much more complex fuel mixtures.

 

5. Database of Geochemical Kinetics of Minerals and Rocks
Ronghua Zhang, Shumin Hu, Xuetong Zhang and Yong Wang
Open Research Laboratory of Geochemical Kinetics, Chinese Academy of Geological Sciences, Institute of Mineral Resources, China

Data of reaction rates of minerals and rocks in waters at high temperatures T and high pressures P are important in understanding the water -rock interactions in lithosphere, and in dealing with the pollution of ground water and deep buried nuclear wastes. Reaction rates have been measured experimentally in the T range 25 to 300 °C and at various pressures. A few kinetic experiments of the mineral dissolution were performed at T above 300 °C and P higher than 22 MPa. Experiments were usually carried out using flow reactors. As operating a continuous stirred tank reactor CSTR reactor, steady state dissolution rates r (mol.sec^-1m^-2) were computed from the measured solution composition using

where

stands for the molar concentration difference between the inlet and outlet of the ith species in solution, F represents the fluid mass flow rate, v_i refers to the stoichiometric content of i in mineral, s is the total mineral surface in the reactor (m²). As operating a flowthrough packed bed reactor PBR, mineral particles were put inside the vertical vessel. Within the PBR, a transient material balance in a column at length Z gives:

This model characterizes mass transfer in the axial direction in terms of an effective longitudinal diffusivity D_L that is superimposed on the plug flow velocity U. The length Z and U have been known. As measured the residence time distribution function of the flow system, we can figure out the D_L. If the boundary condition and initial condition are well known, then, the dissolution rate of the mineral is derived from the following mass balance expression for the concentration of the ith solute in a reactor cell:

where C_i is the concentration of ith species, t is the average residence time, and V is the solution volume in the pressure vessel (ml).

Recently, we measured a lot of mineral dissolution rates (carbonate, fluorite, albite, zeolite, actinolite etc.) in aqueous solutions at high T and P above the water critical point, and found the fluctuation of reaction rates occurs as crossing the critical point. And also we collect reaction rate data in the literature. We performed the geochemical kinetics data base. It includes r (rate law, rate constant k, activation energy E_a, chemical affinity A etc.), the surface nature (s, surface modification… ), t (contact time, accumulation time… ), mineral characters (composition, structure, occurrence, etc.), reaction system, hydrodynamic and physicochemical conditions, analytical method and equipment. The rate law is:

where R_net is the net rate of reaction, k₊ is the rate constant of the forward reaction, a_i is the activity of species i in the rate determining reaction raised to some power m. Others are included, e.g., incongruent dissolution, non-linear dissolution rate, non-linear dynamics in the reaction system (if happened). This data base will also provide simulation models in predicting the water/rock interaction in nature.

 

Track IV-A-1: Current Trends and Challenges in Development of Engineering Materials Databases Chair: Aleksandr Jovanovic, MPA Stuttgart, Germany This session will provide an overview of some major issues related to performance, service and use of engineering materials databases, from the viewpoint of users and developers. The aspects of interest are, e.g., use of rapid prototyping, usability (ie. user friendliness), availability (stand-alone, LAN, Internet/Intranet), safety, reliability, etc. In particular, the issue of integrated, distributed and web-oriented databases and data warehouses will be considered. Most of these aspects require different solutions so the optimum one must be found in each case. Related issues to be addressed include: 1. Measuring performance, service and use of software databases and data warehouses; 2. Internet and intranet databases, including implications of the technology for the 'contents' (i.e. materials data) and for the users; 3. Data vs. information vs. knowledge in engineering materials databases and data warehouses - including e.g. databases of case histories, documentation, etc.; 4. Integrated data assessment for production of higher level information - e.g. automatic definition of material laws based, e.g., on stored materials data and conventional statistics; 5. Use of intelligent methods (neural networks, machine learning, case-based reasoning, fuzzy clustering, etc.); 6. Data consistency, quality/reliability, quality assurance, certification etc. in distributed systems; 7. Future trends. Practical applications of interest would be: large materials databases, European and international engineering materials databases, intelligent databases, corrosion/fatigue/creep databases, material testing databases, databases of certified materials data, Internet databases, materials databases in technology transfer, etc.

1. Development of a Large System of Clustered Engineering Databases for Risk-Based Life Management
A. S. Jovanovic, MPA Stuttgart

The paper describes the development of complex databases system comprising currently more than 30 single databases containing data needed for the risk-based life management of components in industrial plants. The system provides basis for the development of a new European guideline in the area of risk-based life management (RBLM), inspection (RBI) and maintenance (RBIM). Full-scale application of the concepts proposed by the guideline is essentially possible only if the issue of maximum use of available data (and consequent minimization of the need to acquire further data!), and only a modern, comprehensive, but flexible database system can provide the required solution.

The database cluster is organized as a data warehouse satisfying the needs to: (a) work in highly distributed environment, both on the developers' and on the users' end; (b) work with constantly changing database structures, updated/changed at the level of the database administrator (not developer!); (c) share common tolls and tasks across all the databases (e.g. graphics, statistical evaluation, application of data mining tools, etc.); (d) assure linking and possible integration of existing databases of older generation; (e) assure transportability of the system over a wide range of operating systems.

The paper also shows how the principles of RBLM are practically applied in a European power plant, including the implementation aspects in the "non-ideal situation" (lack of data, uncertainties, need to combine experts' opinions with results of engineering analysis, etc.).

 

2. Open Corrosion Expertise Access Network
W.F. Bogaerts, University of Leuven - Materials Information Processing & Corrosion
Engineering Labs, Belgium
H.A. Arents, Information Architects group, Flemish Regional Government, Belgium
J.H. Zheng, University of Leuven - Materials Information Processing & Corrosion
Engineering Labs, Belgium
J. Hubrecht, University of Leuven - Materials Information Processing & Corrosion
Engineering Labs, Belgium
R. Cottis, UMIST - Corrosion and Protection Centre, UK

The paper will describe concepts and results from the European Commission supported "OCEAN" project (Open Corrosion Expertise Network), of which the first phase is about to be finalized during 2002. The objective of this project is the design and implementation of an open, extensible system for providing access to existing corrosion information. This will be achieved through a network of interested data providers, users and developers. Where available, existing standards and technologies will be used, with the partners developing informatics and commercial protocols to allow users single-point access to distributed data collections.

One of the major difficulties of corrosion engineering is the multi-dimensional nature of the corrosion problem. A very large number of alloys are available, and these may then be exposed to an almost infinite range of environments. Thus, although many thousands of corrosion tests have been performed and numerous papers published, it remains difficult for the individual corrosion engineer to bring together the information that is relevant to a specific situation. To some extent this problem has been tackled by centralized collections of corrosion data and abstracts. However, these are limited to published information, and tend to be rather inaccessible to potential users. The latter problem relates partly to the dedicated user interfaces that are typically used with these data collections, and partly to the commercial necessities of ensuring a reasonable return for the information providers.

The OCEAN project aims to overcome these limitations through the development of open protocols for locating, paying for, and obtaining corrosion information. In this context 'information' is used very generally, and the OCEAN system is intended to cover all sources of corrosion information including large centralized data collections, individual data collections from research projects, human expertise distilled into books and expert systems, computer-assisted learning texts, multimedia resources and access to human experts. The nominated partners in the project include representatives of several categories of information providers and users, with interest groups allowing additional organizations to participate in the project. It is a specific objective of the project that OCEAN will be open; open to all information providers to offer information, and open to all data users to obtain information. At the same time the commercial value of information will be recognized through commercial protocols, and partners in the project have particular expertise in funds transfer and electronic information systems (publishing).

The detailed specification of the OCEAN system has been one of the first tasks of the project, and the approach is based on World-Wide Web technology. The core of the OCEAN system will be an intelligent database and re-director that accepts queries in a standard form and then directs them to OCEAN data sources that are registered as having information that may be relevant to the query. The data sources respond with the data requested (or a null return) to the originator of the query. For the initial phase of the project a simple query engine is used to construct correctly formatted queries from user input, and to assemble a single response from the returns from data sources. However, it will also be possible for users to issue queries directly to the OCEAN re-director, or for alternative query engines to be used. This will allow more intelligent front-ends to be developed in due course to support less expert users, or to act as software agents for experts.

 

3. Use of Database Technology for Saving and Rescuing of Perishing Engineering Data and Information In Eastern Europe
L. Tóth, Bay Zoltán Institute for Logistics and Production Systems

One of the driving forces in the development of engineering science are relating to the failures took place in different engineering areas. That is why the results of the failure analysis are representing a high value of worth. Due to the development of the information technology these "local worthies" could become a tool for general access. It is obvious that the results of failure analysis contains always that information which are related to that staff where the case took place, but they contains also information for general using, which support the "thumb rule" of "learning from failures". Relating to the Central and Eastern European countries many "engineering data" (including the material data and failure case studies as well) represents only the "local worth". It is caused by minimum two facts. One of them is relating to the later application of the information technology tools for saving and rescuing of perishing engineering data and information, the other is relating to the attitude of the engineering communities in these countries. Generally it can be said that the responsible specialists for the failure case studies are belonging to the middle or aged generation having the attitudes of the 1965-75 year's of these countries. It means that this generation is not familiar with the possibilities of information technology and the failure cases are regarded as "internal business" for them. Having the new generation's ability to the modern information technology tools these obstacles can be overcome. The best solution seems to be the creation of the Internet technology based national failure case studies warehouses. This database contains on the one hand the open and general information about the failure case studies and the other hand the "teaching aids" related to different type of failures including the methodological procedures of examination of the failures. Having the national case studies databases they can be joined into the network. It can only be effective way to realise it if minimum two criteria are fulfilled. One of them is relating to the unified database structure, the other is to the national language. The uniform database structure and the pilot system have to be developed by using "centralised support" (EU R&D support in Europe, or the support of the insurance companies, etc.). A Hungarian initiative will be presented which contains pp. 400 failure case studies.

 

4. The Background and Development of MatML, a Markup Language for Materials Property Data
E.F. Begley and Charles Sturrock, National Institute of Standards and Technology, USA

MatML is an extensible markup language (XML) for the management and exchange of materials property data. Launched in October 1999 and coordinated by the National Institute of Standards and Technology, an agency of the U.S. Department of Commerce, the MatML project has drawn upon the expertise of a cross-section of the international materials community including private industry, government laboratories, universities, standards organizations, and professional societies. The background and development of MatML will be described and will include a discussion of its features and its relationship to other scientific markup languages.

Track IV-B-1: Toward Interoperable Materials Data Systems Chair: Yoshio Monma, Kochi University of Technology, Japan There has been growing concern for the interoperability of factual databases in the materials database community. In order to have interoperability in the heterogeneous environment of the Internet/Intranet, we need a mechanism for sharing materials information that is not dependent upon computer systems and networking. Currently, the idea of using XML-DTD for the description of materials data is welcomed internationally. Two major activities may be identified: MatML in USA and Europe and NMC's (New Material Center) XML-DTD in Japan. Using XML/Java which supposedly allow platform independence on computer systems in development and operation, some advanced materials databases have achieved success toward being truly interoperable. This session is intended to be a natural sequel to the June 2001 MatML Workshop held at NIST (USA) and cosponsored by the VAMAS TWA 10 (Computerized Materials Data). In this Session we want to exchange ideas and experiences in building and using materials data systems intended to be interoperable in the WWW environment.

1. Requirements for Access to Technical Data -- An Industrial Perspective
Timothy M. King, LSC Group, Tamworth, UK

The ultimate objective of any collaborative venture is to share understanding. Such collaboration is the fundamental basis for all social activity. The modern-day challenge is to collaborate across the globe in an environment where change is an ever-increasing factor. The digital information revolution both fuels and offers to alleviate this challenge. However, the "Tower of Babel" remains a highly relevant parable.

Integration of computer systems is a multi-level problem. While integration is increasingly available across the foundation levels of hardware, software, user access and data, semantic integration is rarely on the basis of an explicit, agreed information model. Such models control the representation of data.

XML is now a major tool in the kit of system integrators. In order to control the content of an XML file, the necessary information model is either a DTD (Document Type Definition) or, increasingly, an XML Schema. Organisations are generating large numbers of different DTDs and XML Schemas to address the needs of individual projects.

Creating information models for integration purposes causes a great deal of pain as different organisations meet to agree and define the terminology and required information capability. The XML community is new to this challenge where as the ISO sub-committee TC184/SC4 <http://www.tc184-sc4.org/> has been working for almost twenty years to create (currently) six standards, including ISO 10303 ("Product data representation and exchange" or "STEP").

The ISO/TC184/SC4 family of information standards addresses a wide range of industrial requirements. Mature parts of the standards have delivered real business benefits to various different projects. Some challenges remain in respect of such information standards: deployment in conjunction with project management requirements; facilitation of concurrent systems engineering; adoption by Small to Medium Enterprises; security; intellectual property rights; legacy systems; and integration of multiple sources. Such requirements remain the barrier between the sources of high quality scientific and technical data and the exploitation of such data within industry.

The WWW and other communities have recognised that XML as a single prevalent representation format is not sufficient and a current hot topic is ontologies. Potentially, ontologies offer a different route to integration where unified definitions across the integration levels offer the basis for automated analysis and creation of integration solutions. However, in the short term, "ontology" is a label that is in use in too many different guises and projects such as the Standard Upper Ontology <http://suo.ieee.org/> will require further development before industry is able to effectively exploit the potential power of ontologies.

2. The Platform System for Federation of Materials' Data by Use of XML
Toshio SHUTO, Yutaka OYATSU, Kohmei HALADA, and Hiroshi YOSHIZU
Mitsubishi Research Institute, Inc., Japan

For improvement of materials database as an intelligent foundation, many databases have been developed from wide ranges of materials. However, most of them are built independently for each field of research and are just as a numerical value fact data. In reality very few are realized as a full-scale utilizable database retrieval system. Regarding material database or material data as common property, easy performance of sharing or mutual use of material database is requested along with utilization of non-material specialized field. To respond to this demand, a prototype of platform system to avail mutual use across boundaries in the field of material database was developed.

 

3. XML data-description for data-sharing of material databases
Kohmei Halada, Director of Ecomaterials Center, National Institute for Materials Science, Japan
Hiroshi Yoshizu, Ecomaterilas Center, National Institute for Materials Science, Japan
Toshio Shuto, Science and Technology Research Division, Mitsubishi Research Institute, Japan
Yoshio Monma, Kochi University of Technology, Japan

The activity of VAMAS , an international collaboration of pre-standardization of advanced material based on the agreement of Versailles summit, on XML-based data-description for data-sharing of material-database is introduced. The description consists of Kernel and Modules of each field of materials properties. The Kernel is developed by NIST, USA. The description of Modules are prepared by NMC, Japan and JRC, EU.

The background of the collaboration is followings. Databases of materials data are widely distributed all over the world. However, the common procedure to retrieve and use the data from the distributed database does not exist. For individual databases, guidelines and standardizations have been prepared such as ASTM E49 especially for materials data. In today's computerized era, further development of common or standardized procedures for the data exchange system from the viewpoint of the common platform, on which data can be treated without the regard to the structure of the original database, is required. The objectives of this activity is to clarify the prerequisite for the generic platform for electrical data-sharing systems of materials data. In order to promote the data-sharing system from multi-resources of materials data where each database has its own inherent structure, it is required to prepare the common basis to retrieve, refer, link and utilize the data among them with electrical exchange.

Now the project finished the Phase 0: feasibility study on the electrical data-sharing platform of distributed materials data, and goes into Phase I: Implementations. In the Implementation's phase, trial and testing the prototype of DTD* template for existing database are subjected. (*DTD is written on the assumption of XML chosen as the result of Phase 0) - creation of prototype of DTD for several existing databases

• documentation of DTD from pre-standardized database structure such as MatML
• comparison and testing with retrieval
• clarification of the requisite of the generic DTD structure for material data

By developing this data-sharing system, various properties of materials which stored in different databases can be linked on the generic platform with the standardized template, in order to use for the life-cycle design of products from comprehensive approaches such as DfE (Design for Environment), DfS (Design for Safety), etc., used in industry.

 

4. A Prototyping of Interoperable System for Data Evaluation of Creep and
Fatigue Data

 

Track IV-B-6: Advances in Handling Physico-Chemical Data in the Internet Era (Part 2) Chairs: William Haynes and Peter Linstrom, National Institute of Standards and Technology, USA Modern communications and computing technology is providing new capabilities for automated data management, distribution, and analysis. For these activities to be successful, data must be characterized in a manner such that all parties will be able to locate and understand each appropriate piece of information. This session will focus on characterization of physico-chemical property data by looking at two related areas: (1) the characterization of physical systems to which data are referenced and (2) the representation of data quality. Scientists have often assessed these quantities in the context of the document in which the data are presented, something automated systems cannot do. Thus, it will be important that new data handling systems find ways to express this information by using methods that can be recognized and fully understood by all users of the data. Many challenges are presented in both of these areas: Characterization – A heat of reaction value, for example, may be a simple scalar number but the system to which it applies is potentially quite complex. All of the species in the reaction must be identified, along with their phases, stoichiometry, the presence of any additional species or catalysts, and the temperature and pressure. Representation – Data quality must be expressed in such a manner that all systems handling the data can deal with it appropriately. Data quality can be considered to have two major attributes: (a) the uncertainties assigned to numerical property values and (b) data integrity in the sense that the data adhere strongly to the original source and conform to well-established database rules.