The following are a set of tools for exploring polymer-polymer and polymer-solvent systems by making use of well-known theories or machine learning models.
Each tool yields plots and tables based on the input, and each plot has zoom options by drawing a box, or by selecting options on the bottom left of the plot. Also on the bottom right is the current mouse position.
The Flory-Huggins model is a lattice model of binary polymer blends that can be used to generate phase diagrams.
The app outputs plot the spinodal curve and a binodal as a function of the volume fraction of the volume fraction of the first species if a given Flory-Huggins chi parameter.
The random phase approximation is a mean-field approach to calculate the linear response of a polymer blend following a thermodynamic fluctuation.
The app outputs the structure factor as a function of the wavevector given the degree of polymerization of A and B, statistical segment length of A and B, volume fraction of A and the Flory-Huggins chi parameter.
A deep neural network (DNN) and Gaussian process regression (GPR) model are trained on
the Polystyrene Cloud Points Dataset.
The app outputs predicted cloud point temperatures as a function of volume fraction for polystyrene in select solvents.
Comparisons to experimental data can also be made (if selected) by searching the dataset and selecting data with the closest Mw by input.