The Theoretically Studies On Quantitative Structure-Property Relationships And Decomposition Mechanism Of Polymers

The diversified physical or chemical descriptors of polymers were choosed, used to calculate and predict the property or reaction mechanism of polymers. The quantum chemical descriptors or topology descriptors, which are obtained directly from repeating units or monomers of polymers, are used to predict polymeric properites and generate QSPR models that have good predictive capability, are easy to apply and in consistence with theoretical analysis. Also discussing the reation mechanism from monomers of polymers, the calculated results indicate the mechanism of polymer.Firstly, the QSPR model of crystal growth rates for polyethylene was built by artificial neural network with crystallization temperature (Tc), molecular weight (Mw) and mass fraction (wt.%). The correlation coefficient R of experimental and calculated values for the training set and testing set are 0.9920 and 0.9844 respectively, and the mean errors of regression and prediction are 0.3448 and 0.3304 respectively. The calculated results indicate that the model obtained in this paper can be well used to predict the rates of crystal growth for polyethylene. The ANN method not only is one of the useful tools for polymeric crystallization process study, but also can overcome the shortcomings of the existing model that can not fit for multiple variables.Secondly, Density functional theory (DFT) calculations are carried out for repeating units of polyalkenes, such as polyethylene, polypropylene, polybutene, polydodecylethylene and polyhexene. The calculated results of the molecular orbital energy of the lowest unoccupied molecular orbital ELUMO, the most negative net atomic charge in a molecule q-, and the entropy S are used to predict the dielectric constantε. The QSPR model obtained from the training sets is evaluated externally using the test sets. The correlation coefficient R betweens the predicted values and experiment values is 0.9532, the standard error is 0.00103. The results indicate that the QSPR model constructed using these quantum-chemical descriptors can be well used to predict the properties of vinyl polymers.Thirdly, A set of new molecular descriptors, the thermal energy Ethermal and the total energy of the whole system EHF, which are obtained directly from polyacrylamides monomer structures, are used to predict the glass transition temperature (Tg) values of polyacrylamides and generate a QSPR model by multiple linear stepwise regression analysis, with correlation coefficient R of 0.9598 and the standard error s of 21.7378. Compared with the existing QSPR models, the proposed model requires only two descriptors, which can be obtained by simple calculation and makes it more valuable to predict the glass transition temperature (Tg) values of polymers.Finally, decomposition mechanism of HCl elimination reaction of polyvinyl chloride (PVC) is investigated using density function theory with B3LYP/6-31G* method. The geometries of reactant, transition state and product have been optimized, the viberational frequency, energy and zero point energy (ZPE) were obtained. The transition state of possible reaction pathway was obtained and verified by internal reaction coordinate (IRC) analysis. The result shows that the ionic reaction mechanism of decom position mechanism of HCl elimination reaction of vinyl chloride is a main reactive channel.The produced QSPR models in this paper are proved to be accurate and reliable. Since these quantum chemical descriptors or topology descriptor express all of the electronic or geometric properties of molecules, can be obtained quickly and accurately by calculation, and have clear physical meanings, these QSPR models are valuable for predicting the properties and providing theoretical guidance for polymeric molecular designs and material designs. We studied the decomposition mechanism of HCl elimination reaction of polyvinyl chloride in theory, which can provide guidance for other polymers about their reaction mechanisms.