(545b) Sensor Material Selection and Response Modeling for the Jpl Electronic Nose Using Molecular Modeling Tools

The JPL electronic nose (ENose) consists of an array of polymer-carbon composite sensing films aimed to detect parts-per-million (ppm) to sub ppm concentration levels of environmental contaminants, for Life Support and Habitation space applications. In preparation for an upcoming, long-term (six month) technology demonstration aboard the international space station (ISS), the JPL ENose team is developing a Third Generation ENose. The previous two JPL ENose generations have been trained to detect, identify and quantify more than 20 chemical species at ppm / sub ppm concentration levels. The target compound list for the Third Generation JPL ENose, under development, includes two additional species, Hg and SO2, with a subset of previously detected analytes. The objective of this investigation is to provide theoretical support for polymer sensor material selection for SO2 and Hg detection and sensor response modeling using molecular modeling tools.

The material selection for detecting SO2 and Hg is investigated using quantum mechanics (QM) [the B3LYP and X3LYP flavors of Density Functional Theory (DFT)]. The methodology adopted involves calculating binding energies of SO2 and Hg with common classes of organic structures, such as alkanes, alkenes, aromatics, primary, secondary and tertiary amines, aldehyde, and carboxylic acid. These QM results were used to develop a first principles force field for use in the calculation of interaction energies of SO2 and Hg atoms with various polymers. The binding energy results of organic-SO2 and Hg systems indicate that a polymer candidate for both SO2 and Hg detection would be one containing amine functional groups (primary, secondary or tertiary amine). Other chemical functionalities in the polymer that have strong binding with SO2 are amides, aldehydes, acids.

The sensor response models are developed using two approaches. The first approach based on first principle molecular dynamics, correlates the sensor responses (normalized resistance changes), with the Hansen components of the cohesive energy of the polymer and target analyte as well as the molar volume of the target analyte. The cohesive energy has contributions from electrostatic, dispersion and hydrogen bond terms. The second approach is based on Quantitative Structure-Activity Relationships (QSAR), where we correlate sensor activities with molecular descriptors using Genetic Function Approximations (GFA). The unique QSAR descriptor set combines the default analyte properties (structural, spatial, topological, conformational, and thermodynamic) with descriptors for sensing film-analyte interactions, which describes the sensor response. The QSAR descriptors in the second approach are calculated using Quantitative Structure-Property Relationships (QSPR) and atomistic simulations.

Keywords: Electronic nose, Environmental monitoring, Material selection, Sensor Response Modeling, Molecular modeling