A general quantitative structure–property relationship treatment for dielectric constants of polymers

Abstract

The dielectric constant (ε) of a polymer represents the capacitance of the polymer relative to a vacuum and is important in many industrial applications. Nevertheless, accurate experimental ε values of polymers are often unavailable, and current prediction methods lack universality and suitability. Reported here is the development and testing of a new QSPR (quantitative structure property relationship) correlation of ε for polymers. The entire set of 57 polymers was divided into a training set of 45 polymers and a test set of 12 polymers according to Kennard and Stones algorithm. A nine‐descriptor model, with squared correlation coefficient (R2) of 0.9384 and standard error of estimation (s) of 0.0873, was produced by using the stepwise multilinear regression analysis (MLRA) on the training set. The reliability of the proposed model was further illustrated using various evaluation techniques: leave‐one‐out cross‐validation procedure, randomization tests, and validation through the test set. POLYM. ENG. SCI., 2011. ©2011 Society of Plastics Engineers.

Fluorescent monomers as building blocks for dye labeled polymers: synthesis and application in energy conversion, biolabeling and sensors

“…136 Homopolymers based on N3 and N4 were utilized, amongst several other polymers, for the quantitative structure-property relationship correlation of the dielectric constant (e) for polymers. 137 An1 was copolymerized with N-neopentyl methacrylamide in a free radical polymerization procedure in order to produce structured polymer Langmuir-Blodgett films. 138 Py2 as well as Py5 were also used in labeled PS films for investigations on the influence of nanoscale confinement on the glass transition temperature.…”

Polyurethanes synthetized with polyols of distinct molar masses: Use of the artificial neural network for prediction of degree of polymerization

“…[ 13,16,17 ] Unlike QSPR/QSAR, quantitative structure–property/activity relations), which followed a mathematical rule to calculate a specific property (such as reaction enthalpy) from descriptors attributes (topological, structural or conformational), ANN systems operate based on comprehensive statistical curves. [ 18‐21 ] Maybe the QSPR/QSAR model's main drawback is that the analyses lack design in the strict sense of experimental study. The analyzed data collected may not reproduce the whole attribute space.…”

Prediction of polymer properties using infinite chain descriptors (ICD) and machine learning: Toward optimized dielectric polymeric materials

“…In the past 30 years, numerous attempts have been made to predict polymer properties, where the glass transition temperature may be the most studied one . Studies on the dielectric constant and refractive index have also been reported . Le et al did a comprehensive review on the QSPR modeling for material properties in 2012.…”

“…Even though studies such as the one described in Katrizky's 1996 paper considered the extension of descriptor values for long chains using a numerical treatment, most of the models simply consider a repeat unit with different end‐cap atoms as a complete molecular representation. Monomers with carbons as end‐cap atoms, dimers, and ring‐like dimers have been used as ways of incorporating features of the polymer chemical environment. With these models, two theoretical issues arise: If a descriptor value is an “extensive” property that varies with the scale of the local molecular representation (e.g., the monomer and dimer give different descriptor values), the information contained in this descriptor is contaminated by the choice of the molecular representation.…”