Modeling the temperature dependance of pKa and integration of chemical process models using SPARC
Ayyampalayam, Saravanaraj Nallagounden
MetadataShow full item record
SPARC (SPARC Performs Automated Reasoning in Chemistry) is a computer program developed to mimic an expert chemist in estimating chemical reactivity parameters and physical properties. SPARC strictly uses structural information of molecules to calculate these properties for a broad range of organic compounds. SPARC currently calculates the following chemical reactivity parameters for organic compounds: ionization pKa, molecular speciation in gaseous, aqueous and non-aqueous medium, hydration constants, hydrolysis constants, and tautomeric equilibrium constants. A temperature dependence of pKa model was implemented to enhance the existing pKa model. The temperature dependence of pKa is modeled using Van’t Hoff’s equation. The Van’t Hoff’s coefficients for the different reaction centers are determined by plotting observed pKa versus the inverse of temperature. The slope of the line is the enthalpic coefficient and the intercept is the entropic coefficient. The temperature dependence was determined and modeled for most of SPARC’s pKa reaction centers. However for reaction centers like amine acting as an acid, the temperature dependence was not modeled due to lack of sufficient experimental temperature dependence data.In the natural environment there are many chemical processes acting on a molecule. In order to model the fate of compounds in nature, the chemical process models in SPARC were integrated. The integrated chemical process model includes hydration, tautomer network and molecular speciation models. This integration was challenging because of the enormous increase in the number of calculations that needed to be performed and the amount of data generated. Intelligent filters, based on the reliability of the calculations performed and identification of unproductive paths, were designed and implemented. In the process of implementing cis-trans isomer and chirality information decoders for SMILES string handling in SPARC, the weakness and incompleteness of the SMILES encoding / decoding rules were discovered. A set of rules for encoding / decoding such information in SMILES was developed and implemented in SPARC. These new rules will be published and a request to collaboratively develop a complete standard for the encoding / decoding SMILES will be made.