The Effect of Charge Distribution On RDX Adsorption in IRMOF-10

Friday, November 13, 2009: 1:30 PM
Canal B (Gaylord Opryland Hotel)

Ruichang Xiong, Chemical & Biomolecular Engineering, University of Tennessee, Knoxville, TN
David J. Keffer, Chemical & Biomolecular Engineering, University of Tennessee, Knoxville, TN
Miguel Fuentes-Cabrera, Center for Nanophase Materials Sciences and Computer Science and Mathematics Division, Joint Institute for Computational Science, University of Tennessee and Oak Ridge National Laboratory, Oak Ridge, TN
Donald M. Nicholson, Computer Science and Mathematics Division, Oak Ridge National Laboratory
Andrea Mcihalkova, Computational Center for Molecular Structure and Interactions, Jackson State University
Tetyana Petrova, Computational Center for Molecular Structure and Interactions, Jackson State University
Jerzy Leszczynski, Computational Center for Molecular Structure and Interactions, Jackson State University
Khorgolkhuu Odbadrakh, Department of Physics, West Virginia University
James P. Lewis, Department of Physics, West Virginia University

Quantum mechanical (QM) calculations and classical grand canonical Monte Carlo (GCMC) simulations were performed to test the effect of charge distribution on hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) adsorption in IRMOF-10. Five different methods, including Mulliken Population Analysis (MPA), Löwdin Population Analysis (LPA), Natural Bond Orbital (NBO) Analysis, the electrostatic potential (ESP)[1] and charges from the MM3 potential[2] were used to assign point charges to IRMOF-10 from ground state structures obtained from QM calculations. Classical GCMC simulations of RDX in IRMOF-10 were performed using each set of point charges. As the charge distributions varied, the adsorption loading and interaction potential energies are significantly different. None of the five sets of point charges when inserted into the MM3 potential are able to quantitatively capture the energy of adsorption or its variation as the position of the RDX adsorbate is changed. One of the potential reasons for the discrepancy between the classical and QM calculations is that the partial charges were generated under different conditions than those present in the simulation of RDX in IRMOF-10. The charge distribution of RDX in the adsorbed phase is significantly different than that in the crystalline phase and the charge distribution of IRMOF-10 also changes significantly with the introduction of an RDX molecule into the cage.

References:

[1] R. Babarao and J.W. Jiang, Molecular screening of metal-organic frameworks for CO2 storage, Langmuir, vol. 24, pp. 6270-6278, 2008.

[2] M. Tafipolsky, S. Amirjalayer and R. Schmid, Ab initio parametrized MM3 force field for the metal-organic framework MOF-5, J. Comput. Chem., vol. 28, pp. 1169-1176, 2007.

Extended Abstract: File Not Uploaded
See more of this Session: Molecular Simulation of Adsorption II:
See more of this Group/Topical: Separations Division