Prediction of intrinsic viscosity in polymer–solvent combinations using a QSPR model

In this work, a linear quantitative structure–property relationship (QSPR) model is presented for the prediction of intrinsic viscosity in polymer solutions. The model was produced by using the multiple linear regression (MLR) technique on a database that consists of 65 polymer–solvent combinations involving 10 different polymer. Among the 30 different physicochemical, topological and structural descriptors that were considered as inputs to the model, only eight variables (four variables for the polymer and four descriptors for the solvent) were selected using the elimination selection stepwise regression method (ES-SWR). The physical meaning of each descriptor is discussed in details. The accuracy of the proposed MLR model is illustrated using various evaluation techniques: leave-one-out cross validation procedure, validation through an external test set and Y-randomization. Furthermore, the calculation of the domain of applicability defines the area of reliable predictions.