A new neural network-group contribution method for estimation of flash point temperature of pure components

In the present study, a new collection of 79 functional groups are used to correlate flash point temperature (FP) of pure components. These functional groups construct an accurate neural network-group contribution correlation to estimate flash point of pure components. For developing the model, 1378 pure components of various chemical families are used. Therefore, the model can be utilized to estimate the FP of pure components without any basic limitations. © 2008 American Chemical Society.