Computer-Aided Design Methodology Of Polymers Based On Molecular Dynamics

Polymers are widely used in the manufacture of various kinds of chemical products.At present,polymers are still primarily designed by trial and error method which is expensive and time-consuming.In the past 30 years,computer aided molecular design(CAMD)has made great progress in the fields of refrigerants,industrial solvents,ionic liquids,polymers and so on.However,because of the complexity between the polymer structure and properties,such as the repeat unit structure,molecular weight distribution,branch of the polymer chain length and distribution are important influence on the properties of the polymer.Most researches on CAMD for polymer products still stay in the phase of repeat unit structure prediction,and there are few studies on other factors affecting polymer properties.Therefore,a computer-aided Polymer Design(CAPD)method is proposed in this thesis to simultaneously consider the influence of polymer repeat unit structure and molecular weight.The polymer products are designed by CAPD with more accurate polymer property model.The main content of this paper includes the following aspects:(1)The CAPD model of two-stage step-by-step solution is established.First,the product requirements and target properties of the polymer are determined.Then,the polymer design problem is transformed into a mixed integer nonlinear programming(MINLP)optimization problem.Using analytic hierarchy process(AHP),the objective function is quantified as a linear function of multi properties.Design variables include repeat unit structure and polymer molecular weight.The constraint conditions of polymer performance and structural feasibility are considered in this optimization model.Then,a two-stage algorithm is established to solve the optimization problem.In the first stage,the effect of molecular weight on polymer properties was ignored,and the relaxed optimization in potential polymer repeat units was solved.In the second stage,molecular dynamics simulation was used to predict the properties of polymers with different molecular weight of repeat units(chain length).At the same time,a specific function is regressed to obtain the effects of molecular weight in each physical property,and the optimal polymer structure and molecular weight were obtained by adding regression function into the original MINLP problem.(2)The established CAPD model was applied to the tire rubber polymer case,and the tire rubber structure with high strength,low temperature resistance and good heat dissipation was designed,namely,the four property constraints of cohesive energy density,density,glass conversion temperature and thermal conductivity were considered.In addition,molecular dynamics is used to simulate the properties of the known molecular structures in the results of repeating units solved in the first stage of CAPD.By comparing the simulation results with the actual reference values,the validity and accuracy of the prediction method established by molecular dynamics are proved.(3)In order to further consider the effect of molecular weight of polymer properties,based on the results of the first stage in CAPD,three different repeat unit structures are selected to simulate several properties with different molecular weight,and the function for each molecular structure is regressed,which contains the repeat unit structure and molecular weight influence.Then,the original MINLP problem is solved by replacing the property prediction function obtained by regression.Thus,the solution of phase two is completed,that is,the optimal solution with both the structure of repeating elements and the molecular weight is achieved.