Coray M. Colina , Pennsylvania State University
The development of cost-effective technologies for the efficient capture and storage of CO2 can help reduce the emission of greenhouse gases to the atmosphere and improve the efficiency of processes in the petrochemical industry that rely on these separations. The aim of this project was to develop computational procedures for optimizing the structure of polymers for use in adsorption-based separations. Polymers of Intrinsic Microporosity (PIMs) are a novel class of porous polymer with potential applications in gas storage, separation, and purification. PIMs are disordered materials, but the chemistry of the framework is well controlled compared to other microporous amorphous material, such as activated carbons. As a result, they can have some of the advantages associated with disordered materials, while maintaining functional control at a nanometer scale by knowing precisely the structure and chemistry of the monomers. The general structural scheme of PIMs is one of rigid backbone segments connected by non-linear or non-planar sites of contortion, which allows for a wide range of chemistries and resulting materials. At its core, our research has focused on this type of microporous amorphous polymeric materials. We have developed algorithms for PIM-1 like structures and simulated the CO2 adsorption. Additionally, though not originally planed, we evaluated the capabilities of several cubic EoS and the perturbed-chain statistical associating fluid theory (PC-SAFT). VLE, LLE of demanding binary mixtures with supercritical CO2 were evaluated, including aromatics and dichlorobenzoates (alkyl and semi-fluorinated). We have shown that these models can capture many important features of the complex phase behavior of highly asymmetric systems composed of supercritical CO2 and alkyl (nonfluorinated or semi-fluorinated) 2,5-dichlorobenzoates. The PC-SAFT was also used to predict the solvent swelling extent of vitrinite- and inertinite-rich coals. Here, the predicted swelling trends obtained from the PC-SAFT were comparable to experimental swelling results. This approach may be a promising tool for subsequent solvent–coal interaction predictions.
During the first year of this project we concentrated on generating a realistic model of the polymeric structure. Development of realistic models of complex polymeric systems is a challenging task, because efficient packing at high densities requires sophisticated computational methods. As a first approximation we generated the structure of PIM-1 under the assumption that the framework is rigid due to the sequence of connected aromatic rings in a glassy state. A polymer chain was constructed using a biased growth algorithm by adding repeat units one at a time; this is a suitable computational model, although in practice these polymers are not synthesized by chain growth but by a step polymerization of two monomers. While using the United Atom TraPPE force-field for nonbonded interactions, we found it necessary to perform a detailed study of the partial charge distribution in the monomer. This improved the overall force-field performance, especially near the dioxin (or diamine) ring. Even though quantum mechanics calculations are demanding, they were necessary to include in this project for the generation of a truly realistic monomer. This task was not originally planned, and although time consuming, was crucial for the next steps of the structure generation and adsorption experiments. This was the focus of the second year of this project.
During the NCE period we developed a new 21-step molecular dynamics compression and relaxation scheme, which provides a virtual amorphous sample comparable to experimental samples. Additionally we determined adsorption isotherms via grand canonical Monte Carlo (GCMC) simulations, which facilitated the understanding of the morphological and dynamical properties of PIM-1, and PIM-1 like structures. Despite the complexity of PIM-1 the simulation results demonstrate the effectiveness of our model when compared to experimental data. The simulations predict CO2 uptake in qualitatively agreement with experimental results. However, the calculated adsorption capacity over predicts the experimental observed values by a factor of 2.
The results of this project were presented at several national and international conferences (8), as well as published (5) in recognized journals in the field.
Since receiving support from the PRF, Coray Colina was awarded the Corning Faculty Fellowship at Penn State, served as co-Director of the REU in Soft Materials, and become liaison of the Computational Molecular Science and Engineering Forum of the AICHE, and liaison/advisor of the Graduate Women in Science Nu Chapter. Recently, July 2011, she was the plenary speaker for the International Year of Chemistry celebration in Bogota, Colombia. This national event was broadcasted live to 10 different cities in Colombia. Gregory Larsen has earned his PhD in Materials Science and Engineering, and is currently in a postdoctoral position in the Energy & Geo-Environmental Engineering Department at Penn State. Graduate student Kyle Hart entered the group in the summer of 2010, and plans to defend his candidacy in fall of 2011. Mr. Wai Fong Chan is currently a graduate student in Chemical Engineering at Virginia Polynechnic Institute and State University.