January 2011 - Allocations-admins - mailman3.cels.anl.gov

[LCRC Accounts] Project Request: power-grid-sim
by accounts＠lcrc.anl.gov 31 Jan '11

31 Jan '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Cong Liu Applicant's institution: ANL Applicant's division: DIS, MCS, ES Project Name: power-grid-sim Project title: Novel Power System Operations Methods for Wind-powered Plug-in Hybrid Electric Vehicles Associated funding: LDRD, DOE-OE Other Systems: Science: In this project, the objective is to propose a novel power system operations method, namely a new stochastic unit commitment and contingency analysis, which can address the uncertainty and variability brought by wind and electric vehicles, while maintaining the stability and reliability of the system. The proposed method will make use of a unified high-performance parallel computing framework for solving the stochastic unit commitment problem as well as implementing contingency analysis. These capabilities can be used by system operators to manage the grid. It can also be applied in planning studies to assess the impact of high penetration of wind and PHEVs on the power system. Project description: The task will be seperated into two subtasks. 1.The biggest challenge of contingency analysis for bulk power systems is the limited computational capability for power flow optimization. Due to the nature of power flow computation, a thorough contingency analysis usually has to confront a brute-force enumeration of contingency scenarios. The number of scenarios is astronomically increased if multiple contingencies are taken into account at the same time (N-k). In addition, the extremely expanded scale of power systems also raises the computational cost for each single power flow run. Last but not least, there has been a need for near-real-time contingency analysis since off-line studies cannot reflect the accurate system conditions due to the mismatch between the real operating states and presumed conditions. However, taking advantage of modern monitoring techniques such as phasor measurement units, contingency analysis tends to be implemented in real time. This kind of on-line contingency analysis calls for more advanced algorithms to achieve reasonable accuracy in a timely manner. 2. Stochastic optimization problems are suitable for high-performance parallel computing architectures since a large fraction of the variables of the problem can be decoupled between uncertainty realizations or scenarios. In the unit commitment problem, the number of coupling variables is proportional to the number of power units to be committed and the scheduling horizon. In the particular case of wind power generation, the number of units can increase dramatically. This loosely coupled structure can be decomposed at the model level (outside the optimization solver). Model-level decomposition techniques such as Lagrangean relaxation and Benders decomposition are easy to implement through a parallel computing way. Six members from different divisions at Argonne will work on this project. The project will use open-source solver such as sparse linear system sover, CBC(mixed integer programming) to solve the problems Project URL: Requested allocation: 350000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 31 Jan '11

31 Jan '11

Hello, A change in allocation has been requested: Requester: olson (Robert Olson) Project: RAST Title: Rapid Genome Annotation Technology Description: The RAST server is a genome annotation server developed by the bioinformatics group in MCS that was funded by the NIH Bioinformatics Resource Centers program. Throughout the history of the RAST server, we have made use only of the computational cluster that was funded by NIH for the purpose of hosting RAST. The configuration of this cluster was extensively hand-tuned for the purposes of running RAST. We believe the technology has reached a state of maturity such that we can develop the tools to enable the core computation to execute on a general-purpose scientific cluster like the LCRC. The real test of the success of this work will be to process real genomic data at scale. To this end we will work toward the integration of the RAST server (which provides facilities for the upload of new genomic data, the monitoring of the progress of the pipeline, and tools for online browsing and analysis of the completed annotation as well as downloads of the completed annotation in one of several export formats) with the LCRC as a computational backend. Initially this integration will be manually supervised, but we plan to investigate the technology for a more automated integration. Advances in the RAST technology have decreased the overall computational load on a per-genome basis, but the number of genomes submitted to RAST continues to rise. In 2010 we were successful at porting the core RAST computations to Fusion, but at the same time we brought the more highly optimized computation online on the core RAST server, highly reducing the requirement for the computational offload. In 2011, however, we envision the return for the need for this offload. We plan to further automate the integration of RAST and Fusion, as well as to leverage the Magellan cluster to engineer a family of computational offload facilities for RAST. The RAST project has also made effective use of the Fusion resource for other computations related to the curation of the protein families that underpin the RAST system. We perform detailed all-to-all analyses of the genomes within families of organisms (for instance, all sequenced genomes in the Brucella family of pathogenic bacteria) to determine the precise relationship between the genes in these organisms. Fusion is an excellent platform for such analyses, and we plan to continue such analyses on the other pathogen groups. Current: undetermined amount Justification: Requested: 250000 A specific reason has been given: We have been running a new large-scale computation to compute detailed similarities between the genomes in our database. We are using the results of this data to compute detailed correspondences between genomes in the database. This data, combined with the data we curate on the linkages between annotations and the published literature, will allow us to create highly consistent and accurate annotations across the entire genome database. The computation thus far has consumed around 40K core hours. It is roughly 18% through the total calculation (based on sequence counts), which gives us an estimate of about 224K core hours for the entire calculation. We are requesting an additional 250K core hours in order to complete this computation with some additional overage in the event our estimates are not completely accurate. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Fwd: Re: Negative Allocation Balance
by John Valdes 27 Jan '11

27 Jan '11

Gary Leaf is running low on time on his project "dynamics" and asked (in person) if we could advance 100K to his project. dynamics was initially given 150K, most of which has been used in the past 2 months. Gary said he has more runs to do as does a collaborator, and between them they can probably use up the 100K by midyear. OK to advance the 100K? John ----- Forwarded message from Gary Leaf <leaf(a)mcs.anl.gov> ----- Date: Thu, 27 Jan 2011 14:35:01 -0600 (CST) From: Gary Leaf <leaf(a)mcs.anl.gov> To: John Valdes <valdes(a)mcs.anl.gov> cc: Gary Leaf <leaf(a)mcs.anl.gov> Subject: Re: Negative Allocation Balance John : What do I need to do in order to obtain some positive balance? -2414%)./start.x Notice: Project "dynamics" has a negative allocation balance; your job will run at the lowest priority. 508211.fmgt2.lcrc.anl.gov (fusion-2415%) GKL ----- End forwarded message -----

1 0

Re: [allocations-admins] [LCRC Accounts] Project Request: foam_engine-processe
by John Valdes 25 Jan '11

25 Jan '11

On Tue, Jan 25, 2011 at 10:37:19AM -0600, Bair, Raymond A. wrote: > > It is Applied, and Transportation (like cfd_enginemodeling) Ah, right, forgot about Transportation. OK, I've created this project now. I asked Sibendu to rename it slightly to "foam-EngineProcesses" from "foam_engine-processe" since the latter name got truncated by the 20-character project name limit, and the lack of an "s" at the end would have no doubtly been annoying for him and us in the long run. :) John

1 0

Re: [allocations-admins] [LCRC Accounts] Project Allocation Request
by Bair, Raymond A. 25 Jan '11

25 Jan '11

Seems like a reasonable request. Ray Bair Chief Computational Scientist, CELS Computing, Environment, and Life Sciences Argonne National Laboratory, and The University of Chicago Sent from my iPad On Jan 25, 2011, at 11:41 AM, "accounts(a)lcrc.anl.gov" <accounts(a)lcrc.anl.gov> wrote: > Hello, > > A change in allocation has been requested: > > Requester: ssom (Sibendu Som) > Project: cfd_enginemodeling > Title: Computational Fluid Dynamics Modeling of Petrodiesel and Biodiesel for Compression Ignition engine applications > Description: 1) Finite volume schemes > 2) LES and RANS turbulence models used > 3) Detailed chemical kinetic models for diesel and biodiesel fuels > 4) CONVERGE software is being used - 24 licenses available > 5) Scales well up-to 24 processors > 6) Detailed spray models used for different spray processes > Current: undetermined amount > Justification: > > Requested: 40000 > > A specific reason has been given: > Ran out of core hours due to reservation on 3 nodes for completing paper deadlines. > > This needs to be approved and the final allocation amount decided upon. > > Thank You, > The LCRC Accounts System > _______________________________________________ > allocations-admins mailing list > allocations-admins(a)lists.lcrc.anl.gov > https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 25 Jan '11

25 Jan '11

Hello, A change in allocation has been requested: Requester: ssom (Sibendu Som) Project: cfd_enginemodeling Title: Computational Fluid Dynamics Modeling of Petrodiesel and Biodiesel for Compression Ignition engine applications Description: 1) Finite volume schemes 2) LES and RANS turbulence models used 3) Detailed chemical kinetic models for diesel and biodiesel fuels 4) CONVERGE software is being used - 24 licenses available 5) Scales well up-to 24 processors 6) Detailed spray models used for different spray processes Current: undetermined amount Justification: Requested: 40000 A specific reason has been given: Ran out of core hours due to reservation on 3 nodes for completing paper deadlines. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Project Request: foam_engine-processe
by John Valdes 24 Jan '11

24 Jan '11

OK. Should this be priority 1 or 3? Shashi will be the POC, I assume. What catetgory and subcat: Engineering Applications and Other? Doesn't look like we have an engine and/or fuel-related subcat, but we could create one. Playing devil's (ie, Allocation Committee's) advocate for a moment, :) how is this project different from cfd_enginemodeling? The Science description of both is very similar, other than the fact that cfd_enginemodeling states it will use Converge whereas this one says it will be using OpenFOAM. If the Science goals and PI are the same, however, I would say using different software isn't sufficient reason to request a different project (they should just request more time on their existing project if needed for the OpenFOAM runs). If the science goals of these projects are indeed different, we can approve this one, but then the Science descriptions of this project and cfd_enginemodeling should be updated so that the differences between them is clear. John On Fri, Jan 21, 2011 at 10:39:42AM -0600, Bair, Raymond A. wrote: > This is the team Shashi is working with, and we have been expecting the project request. Since it is less than 100k hours, we would allocate 60k now. Here we assume they will use about 20k by March 31. > [...] > On Jan 21, 2011, at 9:40 AM, "accounts(a)lcrc.anl.gov" <accounts(a)lcrc.anl.gov> wrote: > > > Hello, > > > > A new project on the LCRC cluster has been requested. Please forward > > the information on to the LCRC Allocation sub-committee. > > > > Applicant's name: Sibendu Som > > Applicant's institution: ANL > > Applicant's division: Energy Systems > > Project Name: foam_engine-processe > > Project title: Assessment of OpenFOAM for different Engine-Process Applications > > Associated funding: DCG-LDRD, DOE-Vehicle Technologies Office > > Other Systems: > > Science: This project will use an Eulerian-Lagrangian Computation Fluid Dynamics (CFD) open source code called OpenFOAM to model various internal combustion engine processes like engine full-cycle and gas-exchange, sprays and gaseous jets behavior, nozzle internal flow, mixture formation, combustion and exhaust emissions. The goal is to assess the performance of OpenFOAM for varied engine-process applications. Scalability of OpenFOAM for different applications will also be tested. > > Project description: The project is divided into two main parts: > > > > In the first part, fundamental studies will be performed to establish flame stabilization and lift-off characteristics of diesel and biodiesel. A new surrogate mixture (methyl decanoate + methyl decenoate + n-heptane) fuel will be used for modeling the combustion characteristics of biodiesel. The major motivation is to develop more predictive and robust sub-models for capturing diesel engine processes. Fairly large chemical-kinetic models, LES turbulence models, fine grid resolution facilitates use of more than one node (8 cores) for simulating these processes. > > > > In the second part, the engine full-cycle will be simulated and the effects of the computational grid on the predicted in-cylinder flow at the intake valve closing (IVC) will be analyzed. Furthermore, the mixture formation process of a gaseous fuel direct-injection will be studied. Here again, the use of computational grids characterized by very high resolution (as expected for the analysis of supersonic jets, jet-to-jet interaction and wall-jet impingement) will require parallel computing (1-2 nodes, 8-16 cores). > > > > Project URL: http://www.transportation.anl.gov/ > > Requested allocation: 60,000 > > Justification: > > > > > > The requester has used 0 hours of their initial startup project. > > > > In addition to approving an initial amount, please specify a Category > > and Subcategory for this project. For a list of the current selection > > of approved categories, please see: > > > > https://wiki.lcrc.anl.gov/wiki/Processes/Categories > > > > Once the Allocation committee has approved the project, please go to > > the Project Management page to create it: > > > > https://accounts.lcrc.anl.gov/projects.php > > > > Thank You, > > The LCRC Accounts System > > _______________________________________________ > > allocations-admins mailing list > > allocations-admins(a)lists.lcrc.anl.gov > > https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

[LCRC Accounts] Project Request: foam_engine-processe
by accounts＠lcrc.anl.gov 21 Jan '11

21 Jan '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Sibendu Som Applicant's institution: ANL Applicant's division: Energy Systems Project Name: foam_engine-processe Project title: Assessment of OpenFOAM for different Engine-Process Applications Associated funding: DCG-LDRD, DOE-Vehicle Technologies Office Other Systems: Science: This project will use an Eulerian-Lagrangian Computation Fluid Dynamics (CFD) open source code called OpenFOAM to model various internal combustion engine processes like engine full-cycle and gas-exchange, sprays and gaseous jets behavior, nozzle internal flow, mixture formation, combustion and exhaust emissions. The goal is to assess the performance of OpenFOAM for varied engine-process applications. Scalability of OpenFOAM for different applications will also be tested. Project description: The project is divided into two main parts: In the first part, fundamental studies will be performed to establish flame stabilization and lift-off characteristics of diesel and biodiesel. A new surrogate mixture (methyl decanoate + methyl decenoate + n-heptane) fuel will be used for modeling the combustion characteristics of biodiesel. The major motivation is to develop more predictive and robust sub-models for capturing diesel engine processes. Fairly large chemical-kinetic models, LES turbulence models, fine grid resolution facilitates use of more than one node (8 cores) for simulating these processes. In the second part, the engine full-cycle will be simulated and the effects of the computational grid on the predicted in-cylinder flow at the intake valve closing (IVC) will be analyzed. Furthermore, the mixture formation process of a gaseous fuel direct-injection will be studied. Here again, the use of computational grids characterized by very high resolution (as expected for the analysis of supersonic jets, jet-to-jet interaction and wall-jet impingement) will require parallel computing (1-2 nodes, 8-16 cores). Project URL: http://www.transportation.anl.gov/ Requested allocation: 60,000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: metal_clusters
by accounts＠lcrc.anl.gov 10 Jan '11

10 Jan '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Glen Ferguson Applicant's institution: ANL Applicant's division: MSD Project Name: metal_clusters Project title: Catalytic Activity of Subnanometer Metal Clusters Associated funding: Competitive LDRD Other Systems: CNM Carbon Cluster 200,000 core-hours, ALCF Intrepid ~2,000,000 core-hours, EMSL Chinook ~200,000 core-hours Science: The production of classical and next generation fuels using highly active, selective and environmentally safe catalysts is a major challenge that, if met, would provide one approach to meeting future energy needs. This challenge, along with others, is a strong driving force for developing catalysts that are more effective. Toward this goal, there is a desire for the rational design of catalysts with a high degree of selectivity and activity. An important, recently published, advancement is the use of monofunctional surface supported subnanometer metal clusters that catalyze several important reactions of small hydrocarbons, i.e., the oxidative dehydrogen of propane or separately the epoxidation of propene. These advancements open the door for new class of catalysts, constructed by rational design, with high activity and selectivity at relatively low temperatures. Two important directions to explore that may improve and extend the usefulness of these catalysts are the design of bifunctional catalysts and the effect of the surface support on the catalyst’s reactivity. Our work in these areas will focus on the use of both density functional and ab initio calculations to predict the reactivity of bimetallic subnanometer alloys and how the surface support affects their catalytic activity and selectivity. One problem is the prediction of catalysts to perform the stepwise dehydrogenation and partial oxidation of propane to propylene oxide. Currently, there is no bifunctional catalyst that forms partially oxidized propene starting from propane. The possibility of a bifunctional metal alloy cluster catalyst gives a unique opportunity to accomplish this goal in a stepwise fashion exploiting the reactivity of the two catalytic sites of the alloy. The alumina (Al2O3) surface was shown to have a central role in the reactivity of supported Ag3 clusters with propene to form propylene oxide. Specifically, the surface was a key player in the dissociation of oxygen, which is the first step in the partial oxidation of propene. This surface reactivity indic ates the possibility the surface support is important in other reactions. We will therefore investigate other surface supports to understand their effects on the catalytic activity of subnanometer bimetallic alloys. These calculations will be closely coupled with experimental work to produce and test the predicted catalysts and surface supports. Project description: The catalytic properties of small bimetallic alloys will be investigated systematically to find those that exhibit the desired bifunctional capabilities. Initially, the focus will be on monofunctional properties of different single component clusters for propane dehydrogenation and propylene epoxidation. The metals to be considered for propane dehydrogenation will include Ir, Rh, Co, etc. and for propylene epoxidation will include Au, Cu, etc. The reactivity of the subnanometer metal alloys will be studied using the periodic density functional theory implemented in VASP. The size of the bimetallic alloy clusters will be as small as reasonably possible to avoid problems of intractable computational cost. The size of monofucntional catalyst in previous studies was three to eight metal atoms. The periodic calculations will allow for an understanding of the themodynamics and kinetics of the catalytic reactions on various surfaces including alumina (Al2O3 ), magnesium oxide (MgO), zinc oxide (ZnO) and graphene. Once the monofunctional properties of the single component clusters are understood, we will investigate bimetallic catalysts to investigate reaction pathways involving two reaction steps, i.e., dehydrogenation followed by epoxidation We will also investigate other properties of these catalytic materials. To determine how surface defects sites influence reactivity cluster models of the surface-bimetallic cluster complex and supercell calculations will be used. To assess various density functional methods we will use large basis set CCSD(T) calculations with the NWChem program and Gaussian 03 to obtain very accurate values for activation barriers and reaction energies. We will work closely with collaborators in the Material Science Division at Argonne to experimentally test the predictions of our modeling efforts. The following calculations will be performed. a. We will initially carry out an assessment of common density functional methods compared to the gold standard of single-reference ab initio quantum chemical methods, CCSD(T), which is known to be accurate for a broad range of problems. These calculations will use small molecule (CH4, CH3OH) reactions with gas phase Cu4 and Co4 clusters to calculate activation barriers for C−H, O−H, and C−O bond breaking these results can then be compared to those of other common density functional methods such as B3LYP, RPBE, and PW91. These calculations will use the NWChem and Gaussian 09 software packages. These calculations will allow us to find the most appropriate density functional method that balances accuracy and cost for subsequent, more computationally demanding, calculations. b. We will calculate the reaction pathways for propane dehydrogenation and propylene epoxidation on single component subnanometer metal clusters. This will be done using the nudged elastic band method implemented in VASP. The reactions will be carried out with both gas phase and supported metal clusters. It is known that the supports can serve an important role in the dissociation of oxygen on the surface. We are interested in determining if, or how, these other support surfaces can affect the catalytic activity. To calculate the effects of local surface defects small cluster models that reproduce the surface effects will be used. c. We will use the thermodynamics and kinetics information from the single-component modeling to choose likely bimetallic cluster candidates for bifunctional catalysis. The bimetallic clusters of different compositions will be optimized and used to study the reaction pathways for dehydrogenation followed by epoxidation. The calculations will include the effects of the supports in the same manner as the single component clusters. Project URL: Requested allocation: 400,000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: cluster_catalysis
by accounts＠lcrc.anl.gov 10 Jan '11

10 Jan '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Glen Ferguson Applicant's institution: ANL Applicant's division: MSD Project Name: cluster_catalysis Project title: Catalytic Activity of Subnanometer Metal Clusters Associated funding: Competitive LDRD Other Systems: CNM Carbon Cluster 200,000 core-hours, ALCF Intrepid ~2,000,000 core-hours, EMSL Chinook ~200,000 core-hours Science: The production of classical and next generation fuels using highly active, selective and environmentally safe catalysts is a major challenge that, if met, would provide one approach to meeting future energy needs. This challenge, along with others, is a strong driving force for developing catalysts that are more effective. Toward this goal, there is a desire for the rational design of catalysts with a high degree of selectivity and activity. An important, recently published, advancement is the use of monofunctional surface supported subnanometer metal clusters that catalyze several important reactions of small hydrocarbons, i.e., the oxidative dehydrogen of propane or separately the epoxidation of propene. These advancements open the door for new class of catalysts, constructed by rational design, with high activity and selectivity at relatively low temperatures. Two important directions to explore that may improve and extend the usefulness of these catalysts are the design of bifunctional catalysts and the effect of the surface support on the catalyst’s reactivity. Our work in these areas will focus on the use of both density functional and ab initio calculations to predict the reactivity of bimetallic subnanometer alloys and how the surface support affects their catalytic activity and selectivity. One problem is the prediction of catalysts to perform the stepwise dehydrogenation and partial oxidation of propane to propylene oxide. Currently, there is no bifunctional catalyst that forms partially oxidized propene starting from propane. The possibility of a bifunctional metal alloy cluster catalyst gives a unique opportunity to accomplish this goal in a stepwise fashion exploiting the reactivity of the two catalytic sites of the alloy. The alumina (Al2O3) surface was shown to have a central role in the reactivity of supported Ag3 clusters with propene to form propylene oxide. Specifically, the surface was a key player in the dissociation of oxygen, which is the first step in the partial oxidation of propene. This surface reactivity indic ates the possibility the surface support is important in other reactions. We will therefore investigate other surface supports to understand their effects on the catalytic activity of subnanometer bimetallic alloys. These calculations will be closely coupled with experimental work to produce and test the predicted catalysts and surface supports. Project description: The catalytic properties of small bimetallic alloys will be investigated systematically to find those that exhibit the desired bifunctional capabilities. Initially, the focus will be on monofunctional properties of different single component clusters for propane dehydrogenation and propylene epoxidation. The metals to be considered for propane dehydrogenation will include Ir, Rh, Co, etc. and for propylene epoxidation will include Au, Cu, etc. The reactivity of the subnanometer metal alloys will be studied using the periodic density functional theory implemented in VASP. The size of the bimetallic alloy clusters will be as small as reasonably possible to avoid problems of intractable computational cost. The size of monofucntional catalyst in previous studies was three to eight metal atoms.1-5 The periodic calculations will allow for an understanding of the themodynamics and kinetics of the catalytic reactions on various surfaces including alumina (Al 2O3), magnesium oxide (MgO), zinc oxide (ZnO) and graphene. Once the monofunctional properties of the single component clusters are understood, we will investigate bimetallic catalysts to investigate reaction pathways involving two reaction steps, i.e., dehydrogenation followed by epoxidation We will also investigate other properties of these catalytic materials. To determine how surface defects sites influence reactivity cluster models of the surface-bimetallic cluster complex and supercell calculations will be used. To assess various density functional methods we will use large basis set CCSD(T) calculations with the NWChem program and Gaussian 03 to obtain very accurate values for activation barriers and reaction energies. We will work closely with collaborators in the Material Science Division at Argonne to experimentally test the predictions of our modeling efforts. The following calculations will be performed. a. We will initially carry out an assessment of common density functional methods compared to the gold standard of single-reference ab initio quantum chemical methods, CCSD(T), which is known to be accurate for a broad range of problems. These calculations will use small molecule (CH4, CH3OH) reactions with gas phase Cu4 and Co4 clusters to calculate activation barriers for C−H, O−H, and C−O bond breaking these results can then be compared to those of other common density functional methods such as B3LYP, RPBE, and PW91. These calculations will use the NWChem and Gaussian 09 software packages. These calculations will allow us to find the most appropriate density functional method that balances accuracy and cost for subsequent, more computationally demanding, calculations. b. We will calculate the reaction pathways for propane dehydrogenation and propylene epoxidation on single component subnanometer metal clusters. This will be done using the nudged elastic band method implemented in VASP. The reactions will be carried out with both gas phase and supported metal clusters. It is known that the supports can serve an important role in the dissociation of oxygen on the surface. We are interested in determining if, or how, these other support surfaces can affect the catalytic activity. To calculate the effects of local surface defects small cluster models that reproduce the surface effects will be used.12 c. We will use the thermodynamics and kinetics information from the single-component modeling to choose likely bimetallic cluster candidates for bifunctional catalysis. The bimetallic clusters of different compositions will be optimized and used to study the reaction pathways for dehydrogenation followed by epoxidation. The calculations will include the effects of the supports in the same manner as the single component clusters. Project URL: Requested allocation: 400,000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0