Evaluation of functionalized isoreticular metal organic frameworks (IRMOFs) as smart nanoporous preconcentrators of RDX

Classical molecular dynamics (MD) and Grand Canonical Monte Carlo (GCMC) simulations were used to generate self-diffusivities, adsorption isotherms and density distributions for hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) in five isoreticular metal-organic frameworks (IRMOFs), which varied in the cage size and in the presence and location of amine groups. These simulations were performed at room temperature (300 K) and low pressures (up to 1 ppm RDX). The atomic charges required for MD and GCMC simulations were calculated from quantum mechanical (QM) calculations using two different charge generation methods—Löwdin Population Analysis and Natural Bond Orbital Analysis. Both charge sets show that the presence of amine groups increases the amount of RDX adsorbed. The cage size and the location of amine groups also affect the loading of RDX. The amount of RDX adsorbed is correlated with the energy of adsorption. The activation energy for diffusion of RDX is not positively correlated with the energy of adsorption. The density distributions identify the location of the adsorption sites of RDX-exclusively in the big cage around the metal complex vertices and between benzene rings. In the absence of amine groups on the framework, one of nitro groups on RDX interacts closely with the metal complex. In the IRMOFs functionalized with amine groups, a second nitro group of the RDX interacts with an amine group, enhancing adsorption. With regard to the application as a smart nanoporous preconcentrator, these IRMOFs are found to concentrate RDX up to 3000 times compared to the gas phase, on a volumetric basis. From a simple Langmuir estimation, the selectivity of RDX over butane is up to 5000. The diffusion of RDX is sufficiently high for real time sensor applications. These results indicate IRMOFs can be tailored with functional groups to be highly selective nanoporous preconcentrators.