Prediction of the onset temperature of decomposition of lubricant additives

Due to the requirements of industrial production, many oil additives must be used at a high temperature. It is of great significance to study the thermal stability of lubricant additives. To our knowledge, there is no literature about the QSPR studies of thermal stability of lubricant additives up to now. This paper, for the first time, reports a multiple linear regression model of the onset temperature of decomposition (Tonset) of lubricant additives. The model based on four descriptors was evaluated with the leave-one-out cross-validation procedure to test the internally predictive ability of the model. Subsequently, the model was evaluated externally with the test set. The root-mean-square errors of Tonsets of the training and test sets are 12.7 and 14.2 K, respectively, which are accurate and acceptable, compared to the errors obtained from other models of thermal decomposition temperatures. The results indicate that the QSPR model constructed in this paper can be applied to predict the Tonsets of lubricant additives.