The absence of powerful analytical methods to characterize complex hydrocarbon streams imposes many limitations on how hydrocarbon conversion processes are modeled. Compositional modeling represents an alternative route towards reducing the empiric nature of these models. The purpose of the present work was to evaluate the requirements to build a molecular representation that retains relevant information about reaction paths and reactivities in order to develop detailed kinetic models.The study of several distillate samples confirmed that indeed compositional modeling techniques are capable of generating hypothetical mixtures of representative species that reproduce the bulk physical and chemical properties of the actual petroleum fraction. It was observed also that the most relevant input comes from the analytical methods that provide molecular classes (PIONA and mass spectra), as they establish the most relevant structural attributes that eventually define the types of conversion reactions.However, such techniques are not always available or required,as opposed to routine petroleum analyses. Bulk properties assist in delimiting molecule size and branching degree, which provides information about reactivity distributions. The comparison against detailed gas chromatography data revealed that the synthetic mixture behaves reasonably well in terms of carbon number distributions but exhibits substantial differences at the molecular level. It was identified that the main problem lies in determining the structural possibilities of alkyl substituents. However, considering all structural combinations of alkyl branches certainly is not feasible for kinetic modeling purposes. Therefore, it is suggested to consider a reduced set of chemically relevant species composed of a main “core” structure and simplified alkyl branch configurations, which can be generated efficiently from standard analytical methods and suitable chemical rules.
The absence of powerful analytical methods to characterize complex hydrocarbon streams imposes many limitations on how hydrocarbon conversion processes are modeled. Compositional modeling represents an alternative route towards reducing the empiric nature of these models. The purpose of the present work was to evaluate the requirements to build a molecular representation that retains relevant information about reaction paths and reactivities in order to develop detailed kinetic models.The study of several distillate samples confirmed that indeed compositional modeling techniques are capable of generating hypothetical mixtures of representative species that reproduce the bulk physical and chemical properties of the actual petroleum fraction. It was observed also that the most relevant input comes from the analytical methods that provide molecular classes (PIONA and mass spectra), as they establish the most relevant structural attributes that eventually define the types of conversion reactions.However, such techniques are not always available or required,as opposed to routine petroleum analyses. Bulk properties assist in delimiting molecule size and branching degree, which provides information about reactivity distributions. The comparison against detailed gas chromatography data revealed that the synthetic mixture behaves reasonably well in terms of carbon number distributions but exhibits substantial differences at the molecular level. It was identified that the main problem lies in determining the structural possibilities of alkyl substituents. However, considering all structural combinations of alkyl branches certainly is not feasible for kinetic modeling purposes. Therefore, it is suggested to consider a reduced set of chemically relevant species composed of a main “core” structure and simplified alkyl branch configurations, which can be generated efficiently from standard analytical methods and suitable chemical rules.
การแปล กรุณารอสักครู่..
