(4ef) Chemical Descriptors and Quantitative Structure Activity Relationships for Catalyst and Materials Design

My research involves the use of quantum chemistry methods like density functional theory (DFT) to study processes at the atomic level. Several quantitative descriptors like net atomic charges and bond orders are important for understanding all materials and chemical processes at this level. Other descriptors are important to broad classes of materials. For example, the amount of steric hindrance in transition metal complexes can be quantified by ligand cone angle and solid angle descriptors. The net magnetization associated with each atom is an important descriptor for magnetic systems.

In some cases, an adequate method for calculating one of these descriptors is lacking, and this slows the progress of science. For example, in molecular systems reasonably accurate bond orders are given by the Wiberg bond index array in the natural atomic orbital basis, but an analogous method does not yet exist for periodic systems. Current methods for periodic systems are based on projection of the local density of states (or similar techniques) which give bond populations not bond orders. Consequently, the mention of bond orders is ubiquitous in molecular chemistry but almost entirely absent in the chemistry of non-molecular crystals. An extension of the natural bond orbital (NBO) method to periodic systems would remedy this problem. I believe that such an extension should be possible and am interested in pursuing research along these lines. Recently, I have initiated collaboration with one of the developers of the NBO method to investigate this.

Judging from the more than twenty thousand citations to Shannon's seminal paper[1], ionic radius is another important concept to many researchers. However, ionic radii are currently available for only about half of the known oxidation states of the elements. Recently, I developed an accurate and universal method for computing the ionic radius of any element in a negative oxidation state, which is a step towards solving this problem. Unfortunately, a corresponding method for the positive oxidation states is still lacking.

With Professor David Sholl (my post-doctoral advisor) at the Georgia Institute of Technology, I recently developed the Density Derived Electrostatic and Chemical (DDEC) charge method.[2] DDEC is the first method applicable to both periodic and nonperiodic materials that simultaneously optimizes the net atomic charges to reproduce each atom's chemical state and the electrostatic potential outside the electron distribution in such a manner that constraints are not required to treat systems with buried atoms. DDEC should find widespread use as a general purpose method for calculating net atomic charges. With Professor Sholl, I also developed a dimensionless reaction coordinate for quantifying the relative lateness of transition states.[3] (A transition state is said to be late if it resembles the products more than the reactants.) A descriptor of this type is important for applying the Hammond-Leffler, structure sensitivity, and reactant sensitivity postulates that describe correlations between activation energies and system structure.[3]

My research group will use computational chemistry methods like DFT to study reaction mechanisms and to compute chemical descriptors that are combined with available experimental reaction rates to construct quantitative structure activity relationships (QSARs). These QSARs will be used to identify changes in the material's structure that are predicted to improve performance. These materials will then be synthesized and tested by experimental collaborators working in the areas of homogeneous and heterogeneous transition metal catalysis and advanced materials.

Under the direction of Professors James Caruthers and Kendall Thomson, this type of process was used in my doctoral research at Purdue University to design improved olefin polymerization catalysts.[4,5] These catalysts were homogeneous Ti and Zr complexes containing mixed cyclopentadienyl/aryloxide ligation. QSARs were developed for each of the main reaction steps. A new type of catalyst system was designed that exhibited superior performance due to opportunistic bonding of the ligand to the metal.[5] A more complete description of this work can be found in the abstract of my doctoral dissertation.[6]

I am particularly interested in using these methods to study problems in the area of green chemistry. For example, antifreeze is commonly made from ethylene glycol which has a sweet taste and is toxic to animals. Every year this leads to the death of many pets and other animals that ingest spilled antifreeze. Propylene glycol is a nontoxic alternative, but due to its more difficult synthesis is not as widely used. Computational modeling could be used to look for better propylene glycol synthesis routes. I have some specific ideas along these lines which could be pursued further.

References:

(1) Shannon, R. D. "Revised Effective Ionic-Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides," Acta Crystallograph. 1976, A32, 751-767.

(2) Manz, T. A.; Sholl, D. S. "Chemically Meaningful Atomic Charges that Reproduce the Electrostatic Potential in Periodic and Nonperiodic Materials," J. Chem. Theory Comput., submitted.

(3) Manz, T. A.; Sholl, D. S. "A Dimensionless Reaction Coordinate for Quantifying the Lateness of Transition States," J. Comput. Chem. 2010, 31, 1528-1541.

(4) Manz, T. A.; Phomphrai, K.; Medvedev, G.; Krishnamurthy, B. B.; Sharma, S.; Haq, J.; Novstrup, K. A.; Thomson, K. T.; Delgass, W. N.; Caruthers, J. M.; Abu-Omar, M. M. "Structure-activity correlation in titanium single-site olefin polymerization catalysts containing mixed cyclopentadienyl/aryloxide ligation," J. Am. Chem. Soc. 2007, 129, 3776-3777.

(5) Manz, T. A.; Sharma, S.; Phomphrai, K.; Novstrup, K. A.; Fenwick, A. E.; Fanwick, P. E.; Medvedev, G. A.; Abu-Omar, M. M.; Delgass, W. N.; Thomson, K. T.; Caruthers, J. M. "Quantitative Effects of Ion Pairing and Sterics on Chain Propagation Kinetics for 1-Hexene Polymerization Catalyzed by Mixed Cp'/ArO Complexes," Organometallics 2008, 27, 5504-5520.

(6) Manz, T. A. "Quantitative Structure Activity Relationships for Olefin Polymerization Catalyzed by Ti and Zr Complexes with Mixed Cyclopentadienyl/Aryloxide Ligation," Ph.D. dissertation, Purdue University, 2009. See my homepage at http://www.prism.gatech.edu/~tmanz3/ for a link to the dissertation abstract.