Prediction of the maximum absorption wavelength of azobenzene dyes by QSPR tools

The maximum absorption wavelength (λmax) of a large data set of 191 azobenzene dyes was predicted by quantitative structure–property relationship (QSPR) tools. The λmax was correlated with the 4 molecular descriptors calculated from the structure of the dyes alone. The multiple linear regression method (MLR) and the non-linear radial basis function neural network (RBFNN) method were applied to develop the models. The statistical parameters provided by the MLR model were R2 = 0.893, Radj2=0.893, qLOO2=0.884, F = 1214.871, RMS = 11.6430 for the training set; and R2 = 0.849, Radj2=0.845, qext2=0.846, F = 207.812, RMS = 14.0919 for the external test set. The RBFNN model gave even improved statistical results: R2 = 0.920, Radj2=0.919, qLOO2=0.898, F = 1664.074, RMS = 9.9215 for the training set, and R2 = 0.895, Radj2=0.892, qext2=0.895, F = 314.256, RMS = 11.6427 for the external test set. This theoretical method provides a simple, precise and an alternative method to obtain λmax of azobenzene dyes.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► A simple and precise method to predict λmax of azobenzene dyes was performed. ► Both linear and none linear model was built by 4 calculated molecular descriptors. ► Nonlinear model produced better predictive ability than the linear MLR model. ► We analyzed the structural factor playing an important role in excitation process.