#### Academic Field

Chemistry, Chemical Engineering

#### Faculty Mentor Name

Dr. Karen Downey

#### Presentation Title

Modeling of group 10-based metallo-organic potential catalysts, using DFT

#### Presentation Type

Poster Presentation

#### Abstract

Metallo-organic complexes are catalytically interesting but difficult to make. Their syntheses often involve materials and produce wastes that are particularly toxic. Accurate simulations could hugely reduce time and materials wasted, production of hazardous wastes, and exposure to toxic compounds. To accomplish this, a reliable method of predicting complex geometry, electron distribution, and molecular orbital energies must be established. Current computational methods have limited success in calculating the probable characteristics of many real solvated complexes. A systematic investigation of the catalytically active group 10 metals uses Density Functional Theory (DFT) calculations to model square planar metallo-organic complexes. The chosen complexes incorporate one tridentate pincer ligand and one of seven monodentate ligands of varying electron donating ability. All combinations of these complexes are modeled in vacuum in Spartan 10 using DFT/B3LYP/6-31G to establish a reliable starting point for solvation models and decrease the computational cost of modelling each complex in different solvents. The most common solvation model, SM8, does not allow for successful solution of Pd and Pt complex’ ground state configurations. To address this failure, computational parameters (including computation package, basis sets, and solvation models) are investigated to develop methodologies that would allow for the comparison of anticipated properties of 1^{st}, 2^{nd}, and 3^{rd} row transition metals. Comparisons are made between the SM8, MMFFaq, and SM5 solvation models, B3LYP, M06, and PW91 basis sets, as well as different computational options such as convergence tolerance and the number of cycles allowed. Computational packages compared include Spartan 10 and GAMESS.

#### Keywords

metallo-organic, DFT, Spartan, GAMESS, Density Functional Theory, SM8, solvation, transition metal, catalysis, group 10 complexes

#### Start Date

10-4-2015 2:00 PM

#### End Date

10-4-2015 2:45 PM

#### Location

SERC House of Fields

Modeling of group 10-based metallo-organic potential catalysts, using DFT

SERC House of Fields

Metallo-organic complexes are catalytically interesting but difficult to make. Their syntheses often involve materials and produce wastes that are particularly toxic. Accurate simulations could hugely reduce time and materials wasted, production of hazardous wastes, and exposure to toxic compounds. To accomplish this, a reliable method of predicting complex geometry, electron distribution, and molecular orbital energies must be established. Current computational methods have limited success in calculating the probable characteristics of many real solvated complexes. A systematic investigation of the catalytically active group 10 metals uses Density Functional Theory (DFT) calculations to model square planar metallo-organic complexes. The chosen complexes incorporate one tridentate pincer ligand and one of seven monodentate ligands of varying electron donating ability. All combinations of these complexes are modeled in vacuum in Spartan 10 using DFT/B3LYP/6-31G to establish a reliable starting point for solvation models and decrease the computational cost of modelling each complex in different solvents. The most common solvation model, SM8, does not allow for successful solution of Pd and Pt complex’ ground state configurations. To address this failure, computational parameters (including computation package, basis sets, and solvation models) are investigated to develop methodologies that would allow for the comparison of anticipated properties of 1^{st}, 2^{nd}, and 3^{rd} row transition metals. Comparisons are made between the SM8, MMFFaq, and SM5 solvation models, B3LYP, M06, and PW91 basis sets, as well as different computational options such as convergence tolerance and the number of cycles allowed. Computational packages compared include Spartan 10 and GAMESS.