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: Lynda Soderholm Applicant's institution: ANL Applicant's division: CSE Project Name: MDcounterions Project title: Competing solvent and counter ion coordination and its effect on metal hydration properties: a polarizable molecular dynamics simulation Associated funding: DOE, OBES, Chemical Sciences Other Systems: We currently have access to the French GENCI CINES computer center, with 200,000 core-hours for calendar year 2012. Science: A detailed description of a metal-ion’s coordination environment in aqueous solution, specifically its coordinating ligands, their numbers and their distances, is a necessary prerequisite to the development of a predictive understanding of the system’s physical and chemical behavior. At the core of this chemistry is the understanding of how a specific ionic configuration around the metal takes place, and how the other solution components, including other dissolved cations, counter ions, and the solvent, influence the structure and dynamics of the targeted species. Historically this knowledge has been hampered both by a lack of an experimental probe able to provide the necessary metrical data and by the absence of computer codes with enough atomistic detail to provide the accuracies necessary to distinguish the small differences significantly impacting the chemistry. This situation has recently changed from the experimental perspective with the av ailability of high-energy (> 60 keV) X-rays that can be used in total scattering (HEXS) experiments to probe atom-atom correlations in solutions out to distances of 1 nm or more. This capability to routinely collect such high quality data is currently unique to the APS. Lead by ANL’s Heavy Elements group, analyses of solution HEXS data suggest that there may be an energetically important competition between the solvent and the counter ions in the metal-ion vicinity that is currently ignored by high-level calculations attempting to reproduce solution behaviors. Depending on the nature of the counter ion, it may either enter the first hydration sphere, with or without replacing first-sphere solvent molecules, or remain in the second coordination sphere. The latter case is particularly interesting. Experiments suggest that the primary metal-anion interaction may include a cluster formed with its own structured hydration shell. This would add significant, but perhaps necessary, complications to the calculations because such an interaction would indirectly perturb the first-hydration sphere interactions, eventually inducing a change in the structural and temporal properties of that hydration sphere. This behavior has neither been previously rec ognized experimentally nor studied theoretically. It is currently unclear what factors determine the competition between solvent molecules and counter ions, and which interactions drive the molecular-level structure and dynamics. The calculations are necessarily too large to be applied in the comprehensive manner required to address this problem. A unifying molecular picture of ion hydration and the directing role of anions and their clusters represents a potentially break-through improvement to current knowledge. We request computer time to complement our ongoing experimental efforts to provide a fundamental basis upon which to build predictive capabilities. Project description: In an attempt to explain, on a molecular level, the heuristic ordering of a solution in which lanthanide/actinide cations, solvent molecules and anions are present, we will employ molecular dynamics modeling to simulate HEXS data. We will use the POLARIS(MD) classical molecular dynamics code developed by our collaborating theorist in Paris (M. Masella, CEA Saclay), in which we have developed the parameters of a many-body polarizable force-field model. This force-field is more sophisticated than common polarizable force-fields as it accounts for all the subtle many-body interactions taking place in a solute-solvent-anion system, including hydrogen bonding and covalent interactions, the latter of which is in the form of a charge-transfer term. The force-field parameters are adjusted to reproduce highly accurate quantum chemical data on representative hydrated clusters. Working together with our collaborative experimentalists, we have chosen to focus on f-ion bromide solutions. In the lanthanide series, we will initially focus on La(III), Eu(III), Er(III), and Yb(III), while our actinide element sets consist of thorium(IV) and curium(III). Much of the experimental data is already available. Parameters for Ln/An-water, water-bromide, and anion-bromide interactions have been developed, and are available for running nanoseconds molecular dynamics simulations with periodic boundary conditions for various solvated models: 1/ for a given Ln/An element, we wish to investigate the effect of the counter ions by comparing the results of simulations conducted without and with counter ions present in the simulation box. In the latter, the counter ion will be placed in the “second coordination shell” as experiment predict, to start from a structural configuration close to what is observed in solution. 2/ for a given Ln/An-bromide system, we will explore the changes of free energy as a function of the Ln/An-bromide distance (potential of mean-force estimated from the results 10 trajectories at fixed Ln/An-bromide distances), to see if and where it reaches a minimum that would correspond to the formation of contact ion pairs. Our specific focus will be a competition for metal ion coordination between a bare water molecule and a water molecule bound in an anion complex. The MD simulations will be used to generate atomistic pair-correlations that can be directly compared with experiment. Code to transform the MD simulations into HEXS pdf spectra has been written by the ANL group and is currently available. The fundamentally new aspect of this work is that these comparsions can be made, with significant accuracies, out to distances of about 0.8 nm in real space. Such comparisons will play a critical role in assessing the force-fields used in the simulations. Estimated computer usage: Requested computer time. Altogether we wish to study 6 metal ions, with various numbers of counter-ions, summing up to about 50 trajectories. The PMF calculations require 10 trajectories per metal-bromide couple, and various numbers of counter ions, adding up to about 200 trajectories. Altogether the requested time sums up to 600,000 core hours. To test the accuracy of the quantum chemical reference data (MP2-level), we would like to perform a couple of benchmark calculations using the latest coupled-cluster implementations in the NWChem code. Given the size of the system, we expect to run these calculations across several nodes (probably 4). Each calculation may request up 24-48 hours walltime. To be conduct, we request 80,000 core hours. In total the project sums up to 680,000 core hours. We do not plan to run non-parallel jobs. Project URL: Requested allocation: 680 Q1: 200 Q2: 160 Q3: 160 Q4: 160 Justification: Each trajectory described above should be run on a 5 ns time scales to reach the equilibrium of the system. Typically for a simulation box size of a cation/anion system solvated by 1000 water molecules, the calculation requests 60 hours (walltime) on 32 processors, that is about 2000 core hours. The POLARIS(MD) code is currently parallelized with MPI scheme. In its current version it well up to about 32 processors, but its scalability drops down to 50 % when reaching about 100 CPUs. This is why we ideally run these calculations on 32 processors on the FUSION platform. However, with the help of Intel scientists and the French Exascale Research Laboratory (ERL), the code is currently being rewritten in parts to adapt to forthcoming exascale multicore computers. A new version will be ready by the end of September. The requester has used undetermined amount 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