Active Thermochemical Tables

From Java CoG Kit
Jump to: navigation, search
  • Gregor von Laszewski, gregor@mcs.anl.gov
  • Branko Ruscic, ruscic@mcs.anl.gov
  • Deepti Kodeboyina

Our primary role in the CMCS project is the development of active thermochemical tables (Active Tables) that provide a computational representation of the relationships between molecular-scale data and derived thermochemical properties of molecules. We will implement in the tables intelligent decision-making features via automated linear analysis and will devise interactive interfaces to the active tables that enable tests of "what-if" scenarios. Our secondary goal will include assisting in the development of a Multiscale Chemical Science Portal and deployment of services to allow access to existing reference databases such as that at NIST.

Note: The introduction and description of ATcT has been obtained from various inputs and may not represent the current state of affairs. A considerable amount of information from this page is acknowledged to be from Dr.Branko Ruscic.

Contents

Recent: Active Thermochemical Tables: A Vision for Petaflop Thermochemistry

The general concept of Active Tables (AT) will be introduced and elaborated using Active Thermochemical Tables (ATcT) as a concrete example. ATcT are a new paradigm that catapults thermochemistry from the traditional sequential approach into the digital computing arena of the 21st century. The underlying Thermochemical Network (TN) approach, which exposes and fully utilizes the maze of inherent multiple interdependencies that is largely ignored or simplified by the traditional approach with a resulting loss of knowledge content, allows ATcT to provide reliable, accurate, and internally consistent thermochemistry needed in many domains of chemistry. The ATcT approach involves, inter alia, a statistical evaluation of experimental and theoretical determinations that define the TN. This is made possible by redundancies in the TN, such as competing measurements and alternate network pathways. The statistical analysis produces a self-consistent TN, from which the optimal thermochemical values are obtained by simultaneous solution in error-weighted space, thus allowing the best possible use of all the knowledge present in the TN. Besides deriving superior results by fully utilizing the available knowledge, ATcT offer a number of additional features that are neither present nor possible in the traditional approach, such as full propagation of new knowledge to all affected results, hypothesis testing and evaluation, discovery of weak links in the TN that provide pointers to new experimental or theoretical determinations, etc. The constantly growing TN has evolved to the current a set of ~750 species interlinked by ~7,000 determinations, relevant to combustion processes. Its solution already involves unacceptably long execution times on commodity computers, posing an imminent need to transition to Teraflop resources. However, based on current timing measurements, even Teraflop resources will become inadequate once the TN reaches ~2500 species. Projections show that our goal TN, which would contain ~10,000 species and would come closer to satisfying the more general thermochemical needs in chemistry, would have execution times for a single run of the order of a millennium on a commodity computer, and almost two years on a Teraflop machine, but it would take less than a day on a Petaflop machine. While there is room for both improving the procedure and re-optimizing various portions of the current code, which may result in a non-negligible linear factor of speedup, the basic bottleneck is in the underlying ~n4 scaling of the problem, which quickly outruns any linear improvements.


Introduction to Active Tables

A team of researchers led by Dr. Branko Ruscic, have developed a new paradigm for providing accurate, reliable, and internally consistent thermochemical values for a comprehensive range of chemical species - Active Tables. ATcT implements a Thermochemical Network (TN) concept that explicitly exposes the manifold of inherent interdependencies ignored in traditional approaches. The TN (see figure) is designed to incorporate all available experimental and computational data, which, through the collaboratory environment of CMCS, can be subjected to critical evaluation by recognized experts in the thermo-chemical field. The ATcT analysis, also accessible from the CMCS portal, produces a self-consistent TN from which it can generate, on demand, user tailored and thoroughly documented thermochemical tables that optimally exploit all the available knowledge.

Fig 1: Integrated Approach

A stream of important scientific results is already flowing out of ATcT related research. A team of researchers at Argonne National Lab led by Dr. Ruscic have created a Core (Argonne) Thermochemical Network, C(A)TN, which is the primary TN that enables ATcT to extract new thermochemistry, and which currently encompasses over 350 chemical species and 1000 relevant measurements. Using ATcT, the Argonne research team has fully confirmed their recent revision of the enthalpy of formation of the pivotal combustion and atmospheric radical, hydroxyl (OH). Furthermore, they have reduced its uncertainty by a factor of ~6.5, thus removing the thermochemistry of this chemical species from the list of potential sources of uncertainty in current chemical models. This achievement was singled out on the front cover of the abstract compendium of the 2004 Combustion Conference organized annually by Office of Basic Energy Sciences of the U. S. Department of Energy. This also resulted in new ATcT values for the combustion-related radical HO2, and for nitrogen oxides(NOx) that were used by kineticists at Argonne to analyze their latest experiments.

Advantages of Active Thermochemical Tables ATcT

As opposed to convential seqeuntial thermochemistry, Active Tables are based on the Thermochemical Network approach and hence have the following advantages:

  • addressing and correcting the deficiencies of traditional tables and at the same time
  • introducing a number of completely new features (such as: rapid update with new information, what-if tests, weakest link isolation, availability of the full covariance matrix, the sensitivity matrix etc.)


ATcT vs Traditional Tables

Fig 2: Comparison of the uncertainty for OH in different tables

The Argonne research team has fully confirmed their recent revision of the enthalpy of formation of the pivotal combustion and atmospheric radical, hydroxyl (OH). The graph below shows the variation in the traditional values and the value produced by ATcT.

Fig 3: Comparison of the uncertainty for HO2 in different tables


Another study,in kinetics by Howard on the reactions, NO + HO2 → OH + NO2, and its reverse, over respective temperature ranges of 232-1271 K and 452-1115 K, the heat of formation at 298 K for HO2-radicals was determined to be 2.5 ± 0.6 kcal mole-1, implying ΔHf = 3.2±0.6 kcal mole-1. However, the new preferred value, as obtained with the Active Thermochemical Tables approach, is ΔHf = 3.64±0.06 kcal mol-1. The graph below shows the reduction in the uncertainty of the value using ATcT.



ATcT Data Organization

The data for Active Tables is maintained in the form of ascii files. They are organized in Libraries and Notes.

Libraries
are large collections of data,typically generated by committees who oversee and anoint their scientific soundness.
Notes
Users can open and maintain their own mini-libraries (“Notes”), containing their own research results, which can complement and/or supersede the data contained in the public libraries. The users’ data can be kept private, shared within a workgroup, or open to the public.

ATcTdataorg.png

Main Library
The Main Library contains the Core (Argonne) Thermochemical Network that is under development and currently contains ~600 chemical species interrelated by ~3500 determinations. As new data is introduced to this TN, a new set of solutions of the TN is periodically computed, adding the previous state of the library to the permanent archives.
Auxiliary Libraries
Auxiliary libraries (e.g. CODATA Library, Gurvich Library, JANAF Library, etc.) are more static in nature and contain non-networked data needed to reproduce the values in various historical tabulations for ready-reference purposes.

More information on organization of data within ATcT and information on important files can be found in this document.


ATcT Portlet and Web service configuration

ATcT Related Manuals and Documentation

ATcT introduction document
This is an introduction to the Engine of Active Tables, the data organiztion, the various input libraries etc. It is not entirely comprehensive but has been written from a user point of view.
Web Service Deployment
Notes on setting up and deploying the web service on any system.
Laptop Install Manual
This is a short document put together with information on installation of software required for developing the active tables web service/portlet etc.
Basic Commands
A list of basic commands that can be issued to the Active Tables engine.
Active_Thermochemical_Tables/BubblesService
These are notes on maintenance of the web-service on Bubbles

Discussion and Presentations

Salt Lake City Meeting
Presentation at Salt Lake City, UT proposing updates to the Active Tables Service and portlet.
SciDac Presentation
Presentation on updates to ATcT mainly the Network Wizard for the SciDac Program.
ATcT 2 Design
Discussion with the CMCS team on the design of newer service and client for Active Tables.
ATcT uses cases
Discussion with Dr.Branko Ruscic on the overview of scenarios.
Asynch Job Handling
Discussion on integrating the Java CoG Kit Broker with ATcT to allow for asynchronous handling of submitted jobs.

More Information


References

  • G. von Laszewski, B. Ruscic, P. Wagstrom, S. Krishnan, K. Amin, Sandeep Nijsure, R. Pinzon, M. Morton, S. Bittner, M. Minkoff, A. Wagner, and J.Hewson. A Grid Based Active Thermochemical Table Framework. Proceedings of the 3rd International Conference on Grid Computing (Grid 2003), November 2002, Baltimore, USA. las02activetable
  • Kaizar Amin, Sandeep Nijsure, and Gregor von Laszewski. "Open Collaborative Grid Services Architecture (OCGSA)" Proceedings of the W3C !EuroWeb 2002 conference: The Web and the Grid, December 2002, Oxford, UK. las02ocgsa
  • B. Ruscic, R. Pinzon, M. Morton, g. von Laszwski, S. Bittner, S. Nijsure, K. Amin, M. Minkoff, and A. Wagner, "An Introduction to Active Thermochemical Tables: Several "Key" Enthalpies of Formation Revisited," submitted to J. Phys. Chem. A. las04atct-j
  • B. Ruscic, R. E. Pinzon, M. L. Morton, B. Wang, A. F. Wagner, G. von Laszewski, S. G. Nijsure, K. A. Amin, S. J. Bittner, and M. Minkoff, "Further Refinements of the Bond Dissociation Energy in Water and Hydroxyl Radical Using the Active Thermochemical Tables Approach," in Proceedings of the 58th International Symposium Molecular Sectrosctroscopy, Columbus, OH, 16-20 June 2003, p. 178. las03chem
  • Gregor von Laszewski, Branko Ruscic, Kaizar Amin, Patrick Wagstrom, Sriram Krishnan, and Sandeep Nijsure. "A Framework for Building Scientific Knowledge Grid applied to Active Thermochemical Tables" International Journal of High Performance Computing Applications (IJHPCA), volume 17 issue 4, winter 2003, pages 431-447. las03knowledge
  • James D. Myers, Thomas C. Allison, Sandra Bittner, Brett Didier, Michael Frenklach, William H. Green, Jr., Yen-Ling Ho, John Hewson, Wendy Koegler, Carina Lansing, David Leahy, Michael Lee, Renata McCoy, Michael Minkoff, Sandeep Nijsure, Gregor von Laszewski, David Montoya, Carmen Pancerella, Reinhardt Pinzon, William Pitz, Larry A. Rahn, Branko Ruscic, Karen Schuchardt, Eric Stephan, Al Wagner, Theresa Windus, and Christine Yang. A Collaborative Informatics Infrastructure for Multi-scale Science. In Second International Workshop on Challenges of Large Applications in Distributed Environments, pages 24-33, Honolulu, HI, 7 June 2004.las04clade
  • Carmen Pancerella, James D. Myers, Thomas C. Allison, Kaizar Amin, Sandra Bittner, Brett Didier, Michael Frenklach, Jr. William H. Green, Yen-Ling Ho, John Hewson, Wendy Koegler, Carina Lansing, David Leahy, Michael Lee, Renata McCoy, MichaelMinkoff, Sandeep Nijsure, Gregor von Laszewski, David Montoya, Reinhardt Pinzon, William Pitz, Larry Rahn, Branko Ruscic, Karen Schuchardt, Eric Stephan, Al Wagner, Baoshan Wang, Theresa Windus, Lili Xu, and Christine Yang. Metadata in the Collaboratory for Multi-Scale Chemical Science. In 2003 Dublin Core Conference: Supporting Communities of Discourse and Practice-Metadata Research and Applications, Seatle, WA, 28 September - 2 October 2003. las03dc
  • Thomas C. Allison, William Barber, Sandra Bittner, Brett Didier, Michael Frenklach, Jr. William H. Green, John Hewson, Wendy Koegler, Carina Lansing, Gregor von Laszewski, David Leahy, Michael Lee, James D. Myers, Renata McCoy, Michael Minkoff, David Montoya, Carmen Pancerella, Reinhardt Pinzon, William Pitz, Larry A. Rahn, Branko Ruscic, Karen Schuchardt, Eric Stephan, Al Wagner, Baoshan Wang, Theresa Windus, Lili Xu, and Christine Yang. An adaptive informatics infrastructure enabling multiscale chemical science. Poster presentation at Supercomputing 2003, 18 November 2003.cmcssc2003-p
Personal tools
Collaboration and Jobs