| Summary: | 碩士 === 國立高雄第一科技大學 === 環境與安全衛生工程研究所 === 104 === Chemical Reactivity hazard has been reported as one of the main causes of fire and explosion in the industries. The reactivity distributes self-reactivity and compatibility. According to EU-REACH regulation, the self-reactivity is categorized into explosive properties which is Physical and chemical properties. Exothermic onset temperature ( T o ) and decomposition energy (Hd) are important self-reactivity parameters. Although many Exothermic onset temperature ( T o ) and decomposition energy (Hd) prediction models are put forward, but most of them are only for small groups and a small amount of data which make the prediction range very limited. However, certain quantities of samples are used in experiments. when chemicals have unknown toxicity or at high prices, it’s difficult to get their To and Hd through experiments. In this regard, taking reliable methods to estimate the To and Hd of compounds is indispensable. Quantitative structure activity relationship (QSAR) approach has been validated to be an effective method for predicting properties of chemical compounds, and it also has been acknowledged worldwide to be one of the predictive methods for providing hazardous information of chemical substances. EU takes this mode of testing as an alternative in REACH regulation.In this work, the To 137 of N-O compounds and Hd 138 of N-O compounds are collected to build up and validate a QSAR model for predicting the To of N-O compounds. Dragon and CODESSA PRO software are adopted to calculate molecular descriptors for each compound. A modified stepwise regression algorithm is applied to find out molecular descriptors that are highly correlated with the To and Hd of N-O compounds.
|
|---|
