You are here:
Estimation of hydrolysis rate constants for carbamates
Citation:
Patel, J., C. Stevens, AND E. Weber. Estimation of hydrolysis rate constants for carbamates. ACS 2017 Annual Spring Meeting, San Francisco, CA, April 02 - 06, 2017.
Impact/Purpose:
This presentation compares three complementary approaches to estimate hydrolysis rates for carbamates, an important chemical class widely used in agriculture as pesticides, herbicides and fungicides. These QSAR/QSPR models were derived and evaluated with compilations of measured data from the scientific literature and regulatory documents. One or more of these models will be implemented in the Chemical Transformation Simulator (CTS) to estimate environmental transformation rates of organic chemicals.
Description:
Cheminformatics based tools, such as the Chemical Transformation Simulator under development in EPA’s Office of Research and Development, are being increasingly used to evaluate chemicals for their potential to degrade in the environment or be transformed through metabolism. Hydrolysis represents a major environmental degradation pathway; unfortunately, only a small fraction of hydrolysis rates for about 85,000 chemicals on the Toxic Substances Control Act (TSCA) inventory are in public domain, making it critical to develop in silico approaches to estimate hydrolysis rate constants. In this presentation, we compare three complementary approaches to estimate hydrolysis rates for carbamates, an important chemical class widely used in agriculture as pesticides, herbicides and fungicides. Fragment-based Quantitative Structure Activity Relationships (QSARs) using Hammett-Taft sigma constants are widely published and implemented for relatively simple functional groups such as carboxylic acid esters, phthalate esters, and organophosphate esters, and we extend these to carbamates. We also develop a pKa based model and a quantitative structure property relationship (QSPR) model, and evaluate them against measured rate constants using R square and root mean square (RMS) error. Our work shows that for our relatively small sample size of carbamates, a Hammett-Taft based fragment model performs best, followed by a pKa and a QSPR model.
