The spitzer space telescope observation of spectra most likely attributable to diverse and abundant populations of polycyclic aromatic hydrocarbons (pahs) in space has led to tremendous interest in…

The spitzer space telescope observation of spectra most likely attributable to diverse and abundant populations of polycyclic aromatic hydrocarbons (pahs) in space has led to tremendous interest in these molecules as tracers of the physical conditions in different astrophysical regions. A major challenge in using pahs as molecular tracers is the complexity of the spectral features in the 3-20 m region. The large number and vibrational similarity of the putative pahs responsible for these spectra necessitate determination for the most accurate basis spectra possible for comparison. IT is essential that these spectra be established in order for the regions explored with the newest generation of observatories such as sofia and jwst to be understood. Current strategies to develop these spectra for individual pahs involve either matrix-isolation IR measurements or quantum chemical calculations of harmonic vibrational frequencies. These strategies have been employed to develop the successful PAH IR spectral database as a repository of basis functions used to fit astronomically observed spectra, but they are limited in important ways. Both techniques provide an adequate description of the molecules in their electronic, vibrational, and rotational ground state, but these conditions do not represent energetically hot regions for pahs near strong radiation fields of STARS and are not direct representations of the gas phase. Some non-negligible matrix effects are known in condensed-phase studies, and the inclusion of anharmonicity in quantum chemical calculations is essential to generate physically-relevant results especially for hot bands. While scaling factors in either case can be useful, they are agnostic to the system studied and are not robustly predictive. One strategy that has emerged to calculate the molecular vibrational structure uses vibrational perturbation theory along with a quartic force field (QFF) to account for higher-order derivatives of the potential energy surface. Qffs can regularly predict the fundamental vibrational frequencies to within 5 CM-1 of experimentally measured values. This level of accuracy represents a reduction in discrepancies by an order of magnitude compared with harmonic frequencies calculated with density functional theory (DFT). The major limitation of the QFF strategy is that the level of electronic-structure theory required to develop a predictive force field is prohibitively time consuming for molecular systems larger than 5 atoms. Recent advances in QFF techniques utilizing informed DFT approaches have pushed the size of the systems studied up to 24 heavy atoms, but relevant pahs can have up to hundreds of atoms. We have developed alternative electronic-structure methods that maintain the accuracy of the coupled-cluster calculations extrapolated to the complete basis set limit with relativistic and core correlation corrections applied: the cccr QFF. These alternative methods are based on simplifications of hartree fock theory in which the computationally intensive two-electron integrals are approximated using empirical parameters. These methods reduce computational time to orders of magnitude less than the cccr calculations. We have derived a set of optimized empirical parameters to minimize the difference between the cccr QFF energies for the geometries used to build the force field for two molecular ions of astrochemical significance. We have shown that IT is possible to derive a set of empirical parameters that will produce RMS energy differences of less than 2 CM-1 for our test systems. We are proposing to adopt this reparameterization strategy and some of the lessons learned from the informed DFT studies to create a semi-empirical method whose tremendous speed will allow US to study the rovibrational structure of large pahs with up to 100S of carbon atoms.